ELECTROPHYSIOLOGICAL INDEXES OF THE DETECTION AND PROCESSING OF AUDITORY DISTRACTORS MJ Corral Thesis submitted to the Department of Psychiatry and Clinical Psychobiology Faculty of Medicine University of Barcelona for the doctoral degree (PhD) in Psychology in accordance with the requirements of the European PhD Diploma Supervised by Dr. Carles Escera Barcelona, July, 2008 to all the people that participated in our experiments with dedication and interest Acknowledgements One of the most wonderful things that research has given me; it has been the opportunity to meet great people. It has been a long way and I know that you remain in me for ever. Firs of all, I would like to thank Prof. Escera for more that I can express in words. I do not think that anyone has been never so patient with me. Thank you to my colleges and friends Vanessa Carral for let me growing with you, Sílvia Corbera for make understand the closest things, Miriam Cortiñas my twin soul, Judith Domínguez for always showing the best part and Iria San Miguel for all your gorgeous love. Thanks to the new generation: Manu Garcia, Jordi Costa and Paco Díaz. Life is easier close to you. Also, I would like to thank to the people of the Department of Psychiatry and Clinical Psychobiology, from the University of Barcelona. Thanks to Prof. Ana Adan for your confidence, Prof. Mar Ariza for your sincerity, Prof. Imma Clemente for making that all of us feel comfortable, even in the worst moments, Prof. Montse Colell for your excellent human quality and for how much you have taught along all these years, Prof. Mª Ángeles Jurado for your special way of showing the things clearly, Prof. Maria Mataró for your wonderful sensitivity, Prof. Ana Narberhaus for your brightness and clarity, Prof. Montse Pérez-Pàmies for showing us always a different way of living, Prof. Roser Pueyo your delicious sense of humor, Prof. Dolors Segarra for being always here, Bàrbara Segura for your natural kindness and Prof. Josep Ma Serra for your exquisite open mind. To all of you thanks for so good lunch times together and for taking me in from the beginning. Especially thanks to our head of the Department during the last years, Prof. Carme Junqué, because your work makes ours more pleasant. Thanks to the people that belonged to the Department and now is working somewhere else. Thanks to Prof. Helena Almirall, Dr. Rosa Díaz, Dr. Mar Matarín, Dr. David Mataix-Cols, Dr. Dolors Polo and Dr. Elena Yago for i your interest and for letting me learn from you. Thanks to Dr. Rafal Nowak for your infinite kindness and patient. Thanks to Dr. Lluís Fuentemilla for showing always a sincere smile. Thanks to Dr. Josep Marco and Dr. Daniel Linares for drawing closer paths between long distances. Thanks to Prof. Maribel Núñez from the Department of Methodology and Behavioral Sciences, from the University of Barcelona, for your help and, because close to your family, I discovered a beautiful way of live. Sincerest thanks to Júlia Ribot for being always a smart example to follow and for so wise advices. Working abroad has being one of the most exciting experiences in my life. I would like to thank Prof. Erich Schröger for accepting me into his nice group and for being a so great person. Many thanks to Dr. Andreas Widmann, Dr. Urte Roeber, Dr. Nicole Wetzel, Dr. Sabine Grimm, Dr. Dagmar Müller and Dr. Thomas Jacobsen for all your help and warm welcome. Really especial thanks to Prof. Stefan Berti from the Johannes Gutenberg-University Mainz for taking care of me and make me feeling always like at home. Although for short periods, I have the opportunity to work with excellent researchers. Thanks to Prof. Mercedes Atienza and Prof. Jose Luis Cantero for so intense moments, full of wisdom. Thanks to Prof. Francisco Barceló for let me work close to you. Thanks to Prof. Elisabet Tubau and Prof. Jose Antonio Aznar for reducing frontiers. Thanks to Prof. Tomas Ortiz, Prof. Fernando Maestú, Dr. Santiago Fernández, Dr. Alberto Fernández, Dr. Pablo Campo and Dr. Almudena Capilla for everything that I learned in Madrid. I can not forget excellent staff members of the University of Barcelona who resolve with kindness and efficiency many of our daily work troubles. Thank you very much to Pilar Bouzas, Lluís Folch, Josep Ma Ortiz and Marina Romay for being so wonderful professionals. ii Much that I am and I do, it is thanks to my parents. A new live begins for all. Thanks to my sister and his boyfriend for all your support. Thanks to my dear aunt, Maruja, and my cousins, Anabel and Óscar, for taking me in your home. We never forget who missed us. Thanks to my big family CorralFernández for trying to be always together and close. Thanks to my new family Martínez-Clavell for your unconditionally. Many thanks to my friends of Garbuix for your understanding, for being always there and for letting me grow up and learn together. You know how special are for me. Thanks to my friend Pere Barrachina for his special way of encouraging me during all the thesis process. Thanks to Laura Mari for your friendship. Thanks to Marina and Antonio for made me feel so loved. Thanks to Sr. Maties, Sr. Lluís, Sr. Joan, Sr. Franco, Sr. Óscar Pelayo, Sr. Rodríguez and Sr. Manel García for your unplayable lessons of live. I would like to give my sincerest thanks to many people from Vallès who makes me love and respect this beautiful place, where much of this work has been done. Especially thanks to the people of Corró d’Amunt and Marata who make feel one more of the big family. At the end of this travel one of the most wonderful surprises in my life was awaiting. Thanks to my partner, my best friend and my love, JM Clavell, for giving pace when I was desperate, for understanding me always, for awaiting when I was missed and for so many good moments lived together during the end of this period. Finally, my acknowledgment to the Spanish Government (Ministerio de Ciencia y Tecnología: FP2000-5453, Ministerio de Educación y Ciencia: PM99-0167, HA2000-0022, BSO2003-02440, SEJ2006-00496/PSIC, Consolider-Ingenio 2010-CSD-2007-00012) and the Catalan Government (Generalitat de Catalunya: SGR2005-000953) for their financial support. ...thanks to God iii Abstract The occurrence of auditory changes in a stable environment may cause distraction. The pattern of this effect in the present thesis depended on the salience of the feature undergoing auditory change, the temporal distance and the spatial location between the task-irrelevant and task-relevant information. Moreover, the electrophysiological recordings helped to elucidate the cognitive processing underlying behavioral distraction. First, an early and automatic call for attention triggered by mismatch negativity (MMN) was generated by various types of changes, which corroborated its role as a genuine change detector. Second, the effective orienting of attention to the unexpected changes indexed by P3a appeared sensitive to the spatial location of the distractors, suggesting that the orienting of attention could be indeed an attentional spatial switch. And third, the cognitive processing of returning to primary task performance after a momentary distraction associated with the reorienting negativity (RON) component depended on the impaired stage of the ongoing target processing. Taken together, the results shed new light on the mental chronometry of auditory distraction. v List of publications Theoretical reviews Escera, C. & Corral, M.J. (2003). The distraction potential (DP), an electrophysiological tracer of involuntary attention control and its dysfunction. In I. Reinvang, M.W. Greenlee & M. Herrmann (Eds.), The cognitive neuroscience of individual differences (63-76). Oldenburg: Bibliotheks-und Informationsystem der Universität Oldenburg. Escera, C. & Corral, M.J. (2007). Role of mismatch negativity and noveltyP3 in involuntary auditory attention. Journal of Psychophysiology, 21 (34), 251-264. Study I Escera, C., Corral, M.J. & Yago, E. (2002). An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, duration and intensity changes. Cognitive Brain Research, 14, 325-332. Study II Corral, M.J. & Escera, C. (2008). Effects of sound location on visual task performance and electrophysiological measures of distraction. Neuroreport, in press. Study III Corral, M.J., Berti, S., Jacobsen, T., Widmann, A., Yago, E., Schröger, E., & Escera, C. (submitted). Distraction effects with different distractor-totarget intervals: a combined behavioral and event-related brain potential study. vii Glossary of abbreviations ADHD ANOVA DEV DP EEG EOG ER ERPs HR MEG MMN MR NOV RON RT SCD SD SOA STD Attention Deficit Hyperactivity Disorder Analysis of variance deviant (tone) Distraction Potential Electroencephalogram Electro-oculogram Error rate Event-related brain potentials Hit rate Magnetoencephalography Mismatch Negativity Miss rate Novel (tone) Reorienting Negativity Response time Scalp current density Standard deviation Stimulus-onset asynchrony Standard (tone) ix Contents Acknowledgements Abstract List of publications Glossary of abbreviations 1. Introduction 1.1. Auditory distraction paradigms 1.2. Locus of the auditory distraction 1.3. ERPs elicited during auditory distraction 1.3.1. Change detector mechanism and MMN 1.3.2. Attention switch and P3a 1.3.3. Working memory and RON 2. Aims and hypotheses 3. Methods 4. Results 4.1. Study I 4.2. Study II 4.3. Study III 5. General discussion 6. Conclusions 7. References Appendices Catalan summary of the thesis Theoretical reviews i v vii ix 1 2 5 9 11 13 15 19 25 33 33 42 60 89 97 99 109 127 xi Introduction 1. Introduction In everyday life, our senses are constantly bombarded by information reaching our various sensory epithelia, although we can only process a subset of the incoming information at any given moment in time. The adaptive control of behavior requires the integration and coordination of information originating from different input modalities or locations in external and internal space (Eimer & Driver, 2001; Spence, 2002). Unexpected events may catch our attention automatically, allowing an evaluation of the relevance of the new information. This ability is an advantage when a child suddenly crosses the road while we are driving or when we locate a strange smell coming from the kitchen. However, this detector of differences is a real disadvantage if we want to read in a crowded railway station or if we are trying to work while people are coming in and out of our office. Therefore, detecting sudden changes in the environment allows us to react to potentially dangerous events, but if there is no relevance associated to the change it merely distracts us from our focus of interest. Recent empirical evidence stresses the importance of the interplay and coordination of stimulus-triggered (bottom-up) processes with top-down mechanisms as the basis for adaptive behavior (Berti, Roeber, & Schroger, 2004; San Miguel, Corral, & Escera, 2008a). Therefore, in order to be able to react to unexpected changes, the cognitive system needs to maintain its openness or distractibility even when focusing on task demands. On the other hand, the cognitive system must be effective enough to shield us from irrelevant changes in the environment. 1 1.1. Auditory distraction paradigms Various kinds of distraction paradigms have tried to recreate the interference of unexpected task-irrelevant changes to the task at hand in the laboratory setting. In these paradigms (Figure 1), participants are instructed to classify auditory (Schroger & Wolff, 1998a) or visual (Escera, Alho, Winkler, & Naatanen, 1998) stimuli while ignoring concurrent, task-irrelevant auditory information. In the auditory-auditory version of the distraction paradigms, participants are instructed to concentrate on a particular dimension of the auditory stimuli, i.e., duration, while ignoring any other aspect of the auditory input, i.e., a task-irrelevant change in, for example, frequency. The sequence is arranged so that in random order, half of the stimuli have a particular duration, while the other half are longer, and the participant is instructed to press the response button corresponding to the duration of the stimulus. In a few of the trials, the standard frequency is slightly increased or decreased, and this task-irrelevant frequency change also yields a behavioral cost in the duration discrimination task: participants respond more slowly and make more errors in these “deviant” trials (Schroger & Wolff, 1998a,b; Schroger, Giard, & Wolff, 2000). Moreover, the paradigm provides reliable and replicable measures of distraction (Roeber, Berti, & Schroger, 2003). and It has proved its usefulness for exploring attention in children (Wetzel, Widmann, Berti, & Schroger, 2006; Wetzel & Schroger, 2007a; Wetzel & Schroger, 2007b), even in children as young as five/six years of age, using a modified version with animal sounds instead of pure tones, (Wetzel, Berti, Widmann, & Schroger, 2004). 2 Introduction In the auditory-visual version of the distraction paradigm, participants are instructed to classify visual stimuli into two particular categories, as fast and as accurately as possible, and to ignore the concomitant auditory stimuli, which are presented one at a time preceding the visual stimulus, usually with an stimulus-onset asynchrony (SOA) of 300 ms, onset-to-onset. These taskirrelevant sounds are manipulated so that the “standard” stimulus (occurring in 80% of the cases) is occasionally and randomly replaced by a "distractor", i.e., a stimulus slightly higher/lower (“deviant”) in frequency or by a single environmental (“novel”) sound (i.e., telephone ringing, a glass breaking, or a drill). In the original version of the task, participants were instructed to classify digits into odd and even categories (Alho, Escera, Diaz, Yago, & Serra, 1997; Escera et al., 1998; Escera, Yago, & Alho, 2001; Escera, Corral, & Yago, 2002; Escera, Yago, Corral, Corbera, & Nuñez, 2003; Jaaskelainen, Alho, Escera, Winkler, Sillanaukee, & Naatanen, 1996; Yago, Escera, Alho, & Giard, 2001; Yago, Escera, Alho, Giard, & SerraGrabulosa, 2003), but in subsequent versions of the auditory-visual distraction paradigm the task was modified to classify digits vs. letters (Polo, Escera, Yago, Alho, Gual, & Grau, 2003), or to decide whether the present digit was bigger or smaller than 5 (San Miguel, Escera, Erhard, Fehr, & Herrmann, in preparation). Figure 1. Auditory distraction paradigms. Arrows indicate the target onset. 3 The auditory-visual distraction paradigm has also been tested in tasks using more complex visual stimuli and decisions, such as classifying drawings into animate (animals) or inanimate (objects) categories (Gumenyuk, Korzyukov, Alho, Escera, & Naatanen, 2004), or in deciding whether the color of a figure (a face) was the same or different from that of its surrounding frame (Dominguez-Borras, Trautmann, Fehr, Ehrard, & Herrmann, & Escera, 2008b), or even whether two natural pictures presented simultaneously were the same or different (Dominguez-Borras, Garcia-Garcia, & Escera, 2008c). In all the tasks described so far, the unexpected occurrence of a distracting sound preceding the visual stimulus causes a delay in participants’ responses, “distracting” current task performance. In some studies, the response time increase in deviant trials was accompanied by a hit rate decrease, caused by an error rate increase (Alho et al., 1997; Escera et al., 1998, 2001, 2002; Jaaskelainen et al., 1996). Like the auditory-auditory distraction paradigm, the auditory-visual distraction has been applied to different populations, showing sizeable distracting effects in participants over 8 years of age (Gumenyuk, Korzyukov, Alho, Escera, Schroger, Ilmoniemi, et al., 2001; Gumenyuk et al., 2004), in elderly people without disabilities (Andres, Parmentier, & Escera, 2006), and also in a range of persons with specific disabilities, including children with ADHD (Gumenyuk et al., 2005) or dyslexia (Corbera & Escera, in preparation), adults with closed head injury (Polo, Newton, Rogers, Escera, & Butler, 2002), chronic alcoholism (Polo et al., 2003), or schizophrenia (Cortinas, Corral, Garrido, Garolera, Pajares, & Escera, 2008). Moreover, these clinical studies demonstrated increased distractibility in most disability groups (ADHD, dyslexia, schizophrenia), and even with aging people without disabilities (Andres et al., 2006). 4 Introduction 1.2. Locus of the auditory distraction So far, a number of parameters have been manipulated in order to identify the cause of distraction. It is well established today that the behavioral distraction effects depend on the size (Berti et al., 2004; Jaaskelainen et al., 1999), predictability (Sussman, Winkler, & Schroger, 2003), and channel proximity of the change to the task-relevant information (Schroger & Wolff, 1998a). That is, large and unpredictable distracters such as a telephone ringing, may obviously yield larger distractions than small distracters, which are physically similar to standard auditory background, or than predictable stimuli (e.g., visually cued ones). The magnitude of the distracting effects has been reported to be proportional to the frequency difference between the deviant and the standard tones (Berti et al., 2004; Jaaskelainen et al., 1999) as observed by Schröger (1996). In the same line, larger distraction effects were found for novel than for deviant trials (Escera et al., 1998; 2001), and for meaningful than for non-meaningful novel sounds (Escera et al., 2003). The predictability of deviant sounds was studied by Sussman et al. (2003). In their experiment using the auditory-auditory distraction paradigm, each sound was preceding by a visual cue. In a predictable condition, the visual stimuli indicated the pitch of the tone. In the unpredictable condition, the visual stimuli were randomly paired with the sounds, but did not provide any information on their pitch. Deviant tones occurring in the unpredictable condition elicited clear behavioral distraction. However, deviant tones in the predictable series failed to increase response time. 5 Channel-separation relates to the psychological distance between distractor and target, being larger in the auditory-visual paradigm, where distractor and target are presented in different sensory modalities, and smaller in the auditory-auditory paradigm, where distractor and target are embedded in the same stimulus which could give rise to a unitary perceptual object. Distraction effects can be also observed with tones deviating in features other than frequency, such as tone duration or intensity (Escera et al., 2002), and similar behavioral and electrophysiological effects could be elicited using deviances in intensity (Rinne, Sarkka, Degerman, Schroger, & Alho, 2006), and location and duration (Roeber, Widmann, & Schroger, 2003). In fact, Jankowiak & Berti (2007) presented fixed auditory sequences with high deviant probabilities (one, two or three standard stimuli to each deviant stimuli), in different conditions and participants. Deviant stimuli varied either in pitch, in loudness or in sound source locations, in such a way that the temporal position of the deviant sound was predictable, but not the deviant feature. Deviant stimulus prolonged reaction responses, in comparison to standard tones, when the deviant stimulus was presented every two (p=0.33) or three (p=0.25) standard stimuli, but not when deviant and standard stimulus were alternated (p=0.5). These results suggested that distraction is affected by the probability and the auditory feature channel of the deviant stimulus. The distracting effects in the auditory-auditory distraction paradigm might be accounted for by the moving of attention from one auditory dimension (e.g., task-relevant duration) to another (e.g., a task-irrelevant pitch change) of the same perceptual object, leading to large behavioral effects (Berti & Schroger, 2003; Escera et al., 1998; Escera, Alho, Schroger, & Winkler, 2000; Escera & Corral, 2007; Jankowiak & Berti, 2007; Schroger & Wolff, 1998a,b; Schroger et al., 2000). Less obvious is the stage of visual taskperformance that is interfered with by the occurrence of the task-irrelevant sound change. For instance, Alho et al. (1997) observed that a visual ERP 6 Introduction component generated in extrastriate visual areas, the N1 elicited to visual targets (Heinze et al., 1994), was attenuated when the preceding sound deviated in pitch compared to when preceded by standard sounds, suggesting that early visual target processing was impaired in distracting trials. However, Parmentier, Elford, Escera, Andres, & San Miguel (2008) failed to confirm this suggestion for novel sounds. In their study, visual stimulus identification was made more difficult by masking the visual target with white noise, which, as expected, prolonged response time considerably, but left the magnitude of the distracting effects of novel sounds similar to those of a control condition with mask. They also tested whether the distractor interfered at a later stage of target processing, instructing participants to classify digits into two or four categories. Again, longer reaction times were found for the four category condition than for the two category condition. However, the magnitude of distraction did not differ significantly between conditions. These authors therefore suggested that the distracting effect of the novel sounds occurs “before” the presentation of the imperative visual stimulus. To confirm this hypothesis, Parmentier et al. (2008) quickly (50 ms) flashed a shrinking cross in the time period between the auditory, task-irrelevant stimulus and the visual target, in what they called a “re-capture” condition. Their results showed that the distracting effects of novel sounds were abolished in the re-capture condition, suggesting that the unexpected taskirrelevant change in the auditory environment engaged, and retained, attention in the auditory modality, and that the response time increase observed in the distracting trials would reflect the time necessary to move attention back from the auditory to the visual modality upon the occurrence of a fresh sensory event in the task-relevant modality, i.e., vision. However, as discussed by the authors (Parmentier et al., 2008), this pattern of results could also be explained by the time necessary to move attention between spatial locations instead of sensory modalities, as the visual stimuli were 7 displayed on a screen located in front of the participant, whereas the sounds were presented through headphones. However, how does a distractor interfere to cause impaired performance? May the task-irrelevant stimulus affect any stage of the task-relevant processing or, is there a particularly vulnerable temporal window between the distractor and the target stimulus which determines more distraction? In a previous study, Schröger (1996) showed impoverished performance (hit rate decrease and reaction time increase) when the task-irrelevant stimulus change was presented at 200 ms from the task-relevant tones, whereas no detrimental performance was found when the interval was extended to 560 ms. Later, Escera et al. (2001) also used two different asynchronies between the task-irrelevant sound and the subsequent imperative visual stimulus. Participants were slower to classify the visual stimulus when it was preceded by a slightly higher deviant tone or a novel environmental sound than when the visual target was preceded by a repetitive standard tone. However, no performance differences were found, irrespective of whether the asynchrony between the task-irrelevant sound and the task-relevant visual stimuli was 245 or 355 ms. In the auditory-auditory distraction paradigm, the temporal distance between distractor and target onset was also manipulated. In this task, the distractor onset coincides with the stimulus onset, whereas the time difference between the short and the long stimuli determines the target onset. For example, when participants are to classify short stimulus of 200 ms and long stimulus of 400 ms, the distractor (task-irrelevant change) begins with stimulus onset, whereas the task-relevant target onset occurs 200 ms later (when the stimulus should have finished, if it was a short stimulus). With this paradigm, it has been shown that asynchronies between deviance onset and target onset of 200 ms (Roeber, Berti, & Schroger, 8 Introduction 2003; Roeber, Widmann et al., 2003; Schroger & Wolff, 1998b; Schroger et al., 2000) and 100 ms (Schroger & Wolff, 1998a) are sufficient to cause behavioral distraction. Moreover, Rinne et al. (2006) showed that taskirrelevant decreases or increases in intensity when participants had to discriminate between two equiprobable sounds which differed in pitch, that is, when the distance between the task-relevant and the task-irrelevant feature was 0 ms, also caused distraction. In addition, Roeber, Widmann et al. (2003) found slower responses when infrequent long stimuli (400 ms) appeared in a sequence of repeated short stimuli (200 ms) and participants had to discriminate the localization of the sound (front or left), that is, when the task-irrelevant feature came 200 ms after the task-relevant information. So it seems that task-irrelevant changes presented from -355 ms to 200 ms to the target onset could impair performance. 1.3. ERPs elicited during auditory distraction The recording of event-related brain potentials (ERPs) during the performance of the auditory-visual distraction paradigms described above allows the investigation of the spatio-temporal dynamics of activation of the cerebral network underlying attention control. The typical ERP recorded in this scenario shows a complex morphology, both for the standard and the distracting trials, as it includes auditory and visual responses. However, a simple arithmetic computation, i.e., the subtraction of the responses elicited to standard trials from those elicited to the distracting ones isolates the 9 neuroelectric activation underlying behavioral distraction; for convenience, Escera & Corral (2003) termed this activation pattern the “distraction potential” (DP). The DP shows a characteristic tri-phasic shape, starting with a negative wave followed by a positive one, and ending in a final phase of a more or less sustained negative potential (Figure 2). Each of these waveforms provides a neurophysiological index of what are considered the three main processes involved in involuntary attention control (Escera et al., 2000): 1) the mechanism of attention capture, associated with the mismatch negativity (MMN) and/or to the N1 ERP, 2) the orienting of attention, associated with main the P3a or novelty-P3, and task performance after a 3) the reorienting of attention towards momentary distraction, associated with what is termed reorienting negativity (RON). However, the precise relationship between these neurophysiological phenomena and the cognitive process attributed to them is not entirely clear. In fact, the DP has been recorded without behavioral distraction effects (Munka & Berti, 2006; Polo et al., 2003), as will be discussed in the following sections. Figure 2. The “distraction potential” (DP). Data in the figure are taken from Escera et al. (2001). 10 Introduction 1.3.1. Change detector mechanism and MMN Automatic detection of potentially relevant auditory events, occurring outside the current locus of attention, has been attributed to at least two cerebral mechanisms. One mechanism is activated by sudden onsets of offsets of stimuli, such as a flashing light or an alarm siren, or when a drilling machine or a refrigerator that has been working all the day suddenly stops. Another mechanism is activated by a particular stimulus that violates a neural trace of the preceding homogeneous stimulation, therefore involving a “change” detection process. A naive application of such a change detector mechanism has been used for years by news broadcasters, who, before starting, usually “call” the listeners’ attention by playing a short series of repeating tone pips finishing with another one of longer duration or higher pitch. The first mechanism is based on a neurophysiological reaction to transient increases/decreases in stimulus energy, and has been associated with the auditory N1 (Escera et al., 1998; Giard, Perrin, Echallier, Thevenet, Froment, & Pernier, 1994; Naatanen & Picton, 1987) or at least with some of its components (Escera et al., 1998; Giard et al., 1994). The second mechanism relies on the dynamic modeling of regularity in the acoustic environment, and on the detection of any stimulus change that does not fit in with a neural trace of this regularity. This is therefore a “change-detector” mechanism which has been associated with the generation of the MMN (Naatanen, 1990, 2007; Schroger, 2007; Winkler, 2007), and can be best isolated by using distractor stimuli that differ slightly from the repetitive stimulation (see a fuller discussion of this issue in Schroger & Wolff, 1998a). 11 The behavioral data obtained with the auditory-auditory (Rinne et al., 2006) and auditory-visual distraction paradigms support the existence of these two attention capture mechanisms. Specifically, the distracting effects observed with this latter paradigm in novel and deviant trials display a distinct pattern. Whereas novel sounds cause a long (about 25 ms) delay in the response time to visual stimuli, compared to standard trials, deviant sounds increase response time only slightly (about 5 ms), but there is a noticeable increase in the number of erroneous classifications, as mentioned above. ERP recordings also show this differential effect, with a clear MMN generated to deviant distractors and a combined N1-enhancement/MMN elicited to novel distractors (Alho et al., 1998; Escera et al., 1998). A large body of evidence indicates that both the MMN and the N1 are generated within the planum temporale of the auditory cortex (Alho, 1995; Alho et al., 1998; Escera et al., 2000; Naatanen & Picton, 1987), with additional contributions from prefrontal regions (see Giard et al., 1994 for the N1; Deouell, 2007, for the MMN). However, most of the ‘distraction’ studies described used a pitch change as a deviant feature, i.e., a distractor stimulus differing in its spectral content from the standard sound. Therefore, we cannot conclusively rule out the possibility that the cerebral mechanism leading to attention switching and distraction is associated with the N1 generated by fresh neurons responding selectively to the specific pitch of the deviant sound (Jacobsen & Schroger, 2001; Yago, Escera et al. 2001). Subsequent studies using tones deviating in duration (Escera et al., 2002; Roeber, Berti, & Schroger, 2003), intensity (Escera et al., 2002; Rinne et al., 2006) or location (Roeber, Widmann et al., 2003) helped to corroborate MMN as a genuine change detection involved in involuntary attention switching. 12 Introduction 1.3.2. Attention switch and P3a The P3a waveform, generated with large amplitude to novel stimuli and therefore called "novelty-P3" (see Simons, Graham, Miles, & Chen, 2001, for a discussion on whether the P3a and novelty-P3 can be considered the same ERP component), has been taken in the psychophysiological literature as a cerebral signature of the orienting response (Friedman, Cycowicz, & Gaeta, 2001; Knight, 1984; Squires, Squires, & Hillyard, 1975). Studies using the auditory-visual distraction paradigm have shown that novelty-P3 generation is accomplished in two consecutive phases, each of them involving a different scalp distribution and a different sensitivity to attentional manipulations (Escera et al., 1998, 2001). The first novelty-P3 phase, with peak latency between 220 and 320 ms, has a centrally distributed topography and appears to be independent of attentional manipulations (Escera et al., 1998; (Dominguez-Borras, Garcia-Garcia, & Escera, 2008a; San Miguel, Corral, & Escera, 2008a; see however Dominguez-Borras, Garcia-Garcia, & Escera, 2008b). On the other hand, the second phase of the novelty-P3, occurring between 300 and 400 ms, has a right frontal scalp distribution, and appears highly sensitive to attentional manipulations, as it increases in amplitude when, for instance, participants can monitor the sounds, i.e., in the auditory-visual paradigm, compared with a condition of passive listening (Escera et al., 1998; 2003). In agreement with previous results (Herrmann & Knight, 2001), the scalp-current density (SCD) analysis of the novelty-P3 recorded in the auditory-visual distraction paradigm has shown that novelty-P3 generation encompasses at least five different cerebral regions, engaged in clear spatiotemporal orchestration (Yago et al., 2003). This study suggested that the anterior cingulated cortex was activated first (circa 160 ms), followed by simultaneous activation of the bilateral temporoparietal and the left frontotemporal cortices (around 200 13 ms), to finish with activation of the superior parietal cortex and prefrontal regions (at 300 ms). Whereas the MMN and N1 mechanisms seem to operate automatically, the subsequent brain response, the P3a or novelty-P3, reflecting, according to the most accepted view, the orienting of attention towards unexpected deviant or novel sounds (Escera et al., 1998; 2000; Friedmann et al., 2001; Herrmann & Knight, 2001; Knight, 1984), is largely dependent on top-down factors. The studies reviewed above have shown that the P3a elicited to deviant tones in the auditory-auditory distraction paradigm the novelty-P3 elicited to novel sounds in the auditory-visual distraction paradigm, can increase or reduce their amplitude in parallel with the effects of similar direction at behavioral level (i.e., enhanced or reduced distractibility) depending on the task at hand. This would support the P3a/novelty P3 as the scalp signature of the involuntary orienting of attention, or in other words, one may take the occurrence of a distinct P3a as proof that an effective orienting of attention towards distracting stimuli has taken place. However, this interpretation has recently been challenged by a study by Rinne et al. (2006), who found that behavioral distraction increased as a function of intensity decreases in the absence of any P3a elicitation, contrasting with the results obtained for intensity increases, where behavioral distraction increases as a function of intensity increases were paralleled by similar increases in P3a amplitude. These results call for a reconsideration of the role attributed to P3a generation in involuntary attention models. This is not the only case in which P3a/behavioraldistraction dissociations have been observed using the auditory distraction paradigm. Using its visual version, Munka & Berti (2006) observed the opposite phenomenon: generation of P3a in the absence of any signs of behavioral distraction. Similar results were obtained by (Yago, Corral, & Escera, 2001), who, in a study manipulating parametrically the deviantstandard frequency difference, found that the P3a (and also MMN and 14 Introduction RON) increase linearly as a function of change magnitude, whereas no behavioral effects were observed except in the 10% change condition. These authors even reported a facilitation effect for a condition with 5% frequency change. A further dissociation was observed by Polo et al. (2003), where larger (late) novelty-P3s were observed in a group of chronic alcoholics compared to matched controls, in the absence of any differences in behavioral distraction. Using a very different paradigm without auditory stimulation, either deviant or standard stimuli, Berti (2008b) found that P3a was elicited. The task consisted in performing arithmetic operations on stored numbers. These operations could be performed on a previously processed item or on a new item of the three stimuli presented on the screen. Switch and no switch trials were equiprobable and slower responses were recorded for the switch trials. Moreover, switch trials showed an increased P3a around 300 ms. Therefore, in contrast to the widely held view of the P3a as an orienting attention response or as a distractibility index, the P3a/novelty-P3 may be a signature of the evaluation of the contextual novelty of an unexpected sound, and furthermore, that it may reflect the reconfiguration of a cerebral network involved in updating task-set information for goal-directed action selection (Barcelo, Escera, Corral, & Perianez, 2006). 1.3.3. Working memory and RON As important as the flexibility to direct attention towards unexpected potentially relevant events outside the focus of attention is the ability to 15 return attention back to original task performance after a momentary distraction. It has been proposed that this attentional process is associated with the generation of another ERP component recorded in distraction tasks, known as reorienting negativity (RON; Escera et al., 2001; Schroger & Wolff, 1998b). In fact, in their recordings Schroger & Wolff (1998b) only found a negative waveform subsequent to P3a when their participants carried out a task where the deviant stimuli acted as behavioral distractors, but not when the participants were asked to discriminate these stimuli actively, or when they were instructed to ignore the auditory stimulation completely and to concentrate on an unrelated visual task. Moreover, Escera et al. (2001) argued that in order to indicate the process of reorienting attention back towards main task performance, RON should be time-locked to target stimuli in the task, and not to distracting ones. In their experiment using the auditory-visual distraction paradigm, the asynchrony between the distractor and the visual target was manipulated to 245 or 355 ms, in different conditions. Their results showed that RON peak-latency was about 345 ms, irrespectively of distractor-target asynchrony, i.e., from visual target onset. These results strongly suggested that RON generation reflects the process of returning attention back to primary task performance after a momentary distraction. On the other hand, Escera et al. (2001) observed that RON actually had two different phases or subcomponents, which were dissociated on the basis of their different scalp distribution, and that the late subcomponent was the one time-locked to task-relevant aspects of stimulation, i.e., the visual stimulus. The existence of two RON subcomponents was confirmed in a recent study by Munka and Berti (2006) using the auditory-visual distraction paradigm. These authors found that an early RON component was elicited when the discrimination task had a working memory component, i.e., consisting in making a semantic judgment on the visual stimuli (in fact, a classical odd/even classification). However, when the decision was based on 16 Introduction a physical feature of the stimuli (size or color, in different experiments), only a late RON subcomponent could be observed. On the basis of their results, the authors suggested that the early RON was an electrophysiological correlate of attention refocusing as a working memory process in the sense of a switch of the focus of attention, whereas the later RON seems to be related to a more general aspect of attentional allocation or evaluation after distraction (i.e., rehearsal of instructions or self-motivation). Moreover, a recent study using magnetoencephalography (MEG) found the activation of the primary motor cortex during the RON interval (Horvath, Maess, Berti, & Schroger, 2008), suggesting that, in addition to the frontal activation observed in the ERP studies, the primary motor cortex also contributes to the reorientation of attention to the main task. 17 Aims and hypothesis 2. Aims and hypothesis The general aim of the present thesis was to contribute to our understanding of the neural mechanisms of auditory distraction. Specific objectives and hypothesis are considered separately for each of the three studies, as follows. Study I Escera, C., Corral, M.J. & Yago, E. (2002). An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, duration and intensity changes. Cognitive Brain Research, 14, 325-332. Until this study was published, most ‘distraction’ studies used pitch as deviant feature. Because distractor stimuli differed in their spectral content from the standard sounds, the possibility that the cerebral mechanisms leading to attention switching and distraction was associated to the N1 generated by fresh neurons responding selectively to the specific pitch of the deviant sound could not be conclusively ruled out (Jacobsen & Schroger, 2001; Yago, Escera et al. 2001). Aim: The present experiment sought to investigate whether a genuine change detection indexed by MMN was involved in cross-modal attention switching, i.e., in directing attention involuntarily from the current visual 19 performance to unexpected auditory changes. We used a variant of our auditory–visual distraction paradigm in which the ‘distracting’ sounds were either shorter in duration or softer in intensity, in separate blocks, than the corresponding standard stimuli. An additional condition with frequency deviants was also used. Hypothesis: Because shorter or softer deviant tones are unlikely to activate new fresh neural elements (Naatanen, Paavilainen, & Reinikainen, 1989; Naatanen, Paavilainen, Alho, Reinikainen, & Sams, 1989), observation of behavioral distraction to these sounds will strongly support the involvement of a genuine change detection mechanism in controlling involuntarily the cross-modal direction of attention. Otherwise, if the change detection mechanism was only presented in the pitch change condition and not in the intensity or duration condition, it would have been difficult to maintain MMN as a genuine change detection mechanism involved in involuntary attention. Study II Corral, M.J. & Escera, C. Effects of sound location on visual task performance and electrophysiological measures of distraction. Neuroreport, in press. A critical issue for gaining insight into the neural mechanisms of distraction would be to establish which cognitive process, or processes, during visual task performance are affected by the occurrence of unexpected deviant or novel sounds. Parmentier et al. (2008) showed that the distracting novel sound interfered not during visual stimulus identification or during response selection, but before the target processing. The next open question was 20 Aims and hypothesis whether the distraction effects observed with the auditory-visual distraction paradigm were due to the time necessary to move attention between spatial locations or between sensory modalities. Aim: The objective of this study was to investigate the role of the spatial source of the auditory stimuli in distraction in a visual task-performance. For this purpose, sound location was varied systematically across conditions, while visual target location remained constant. Hypothesis: If the cause of distraction relies only on moving attention between sensory modalities, placing the source of the task-irrelevant auditory stimuli in different spatial locations should not affect the distracting effects of these sounds. However, if a spatial factor is involved in distraction, the greater the distance between the auditory and the visual stimulus sources the larger the distracting effects of novel distracting sounds on visual task-performance, and eventually on the accompanying ERP signs of distraction. Study III Corral, M.J., Berti, S., Jacobsen, T., Widmann, A., Yago, E., Schröger, E., & Escera, C. (submitted). Distraction effects with different distractor-totarget intervals: a combined behavioral and event-related brain potential study. In a previous study, Schröger (1996) showed impaired performance (hit rate decrease and response time increase) when the task-irrelevant stimulus change was presented at 200 ms from the task-relevant tones, but no deterioration when the interval was increased to 560 ms. Later, Escera et al. 21 (2001) also used two different asynchronies between the task-irrelevant sound and the subsequent imperative visual stimulus. Participants were slower to classify the visual stimulus when it was preceded by a slightly higher deviant tone or a novel environmental sound than when the visual target was preceded by a repetitive standard tone. However, no performance differences were found when the asynchrony between the taskirrelevant sound and the task-relevant visual stimuli was of 245 or 355 ms. Moreover, a negative component related to reorienting of attention towards main task performance after distraction (RON) appeared synchronized to the task-relevant information. Other studies have recorded distraction using a different interval between task-relevant and task-irrelevant information yielded distraction (from -355 to 200 ms; Rinne et al., 2006; Roeber, Berti & Schroger, 2003; Roeber, Widmann et al., 2003; Schroger and Wolff, 1998a,b; Schroger et al., 2000). However, the question of whether there is a particular vulnerable temporal window between the distractor and the target stimulus which determines more distraction has not been systematically studied. Aim: The present study addressed two specific questions. First, whether the occurrences of the auditory change at different asynchronies from the target feature onset could affect the magnitude of distraction. Second, whether the reorienting of attention indexed by RON is insensitive to the deviance onset, as in the previous study by Escera et al. (2001). For this purpose, participants were instructed to respond to short and long stimuli and to ignore all other aspects of stimulation. Most of the tones (p=0.88) had the same pitch from the beginning to the end (standard tones, STD) and in a small set of stimuli (p=0.12) slight task-irrelevant pitch changes were introduced (deviant stimuli). Two time points were selected maintaining a similar distance between them of around 100 ms, as in a previous study (Escera et al., 2001). One pitch deviance was presented early, at 50 ms 22 Aims and hypothesis from stimulus onset (early-DEV). The other pitch deviance was presented late, at 150 ms from stimulus onset (late-DEV). Because the short stimuli lasted 200 ms and the long stimuli 400 ms, the discrimination point between them occurred at 200 ms from stimulus onset (the point when the short stimulus finished and the long stimulus continued). So, target onset remained constant through deviant stimuli. Hypothesis: If there is a temporal window that determines more distraction, different behavioral responses should be observed between conditions. Moreover, if RON was a pure index of reorienting of attention, its generation has been related with the target onset independently of the distraction onset (as has been shown in auditory-visual paradigm, Escera et al., 2001). However, if RON generation involves deviant information, peaks at different time windows should be expected. 23 Methods 3. Methods Electrophysiological and behavioral measures were obtained from nondisabled people across all the three studies with the following specifications. Sample Participants were undergraduate students from the University of Barcelona for the first and the second study, and from the Institut für Psychologie I of the Universität Leipzig (Germany) for the third study. Table 1. Sample features N Study I Study II Study III 12 13 12 males 2 7 2 mean age 21 23 21 minimun age 20 19 18 maximun age 23 30 25 All were right-handed, none reported auditory dysfunction and all participated in the study for either course credits or payment. Informed consent was obtained from each participant after the nature of the study had been explained to them. Stimuli Visual and auditory stimuli were presented for the first and second study, while only auditory stimuli were used for the third study. Visual stimuli 25 consisted on the digits from 1 to 8 presented equiprobably during 200 ms on a computer screen. Table 2. Visual stimuli features participant’s distance screen Study I Study II 1.50 m 1.15 m vertical view angle 1.78º 2.00º horitzontal view angle 1.18º 1.25º Auditory stimuli consisted of repetitive standard tones occasionally replaced by slightly deviant or novel sounds. All of them were delivered binaurally through headphones, except for the second study which included five conditions with auditory stimuli delivered through loudspeakers. Table 3. Auditory stimuli features stimuli standard Study I deviant-FRE deviant-DUR deviant-INT standard Study II novel pitch 600 Hz 700 Hz 600 Hz 600 Hz 600 Hz complex sounds (n=60) 1,000 Hz 400 ms Study III early-deviant 1,100 Hz after 50 ms from stimulus onset 1,100 Hz after 150 ms from stimulus onset 200 ms 400 ms 200 ms 400 ms 72 dB 72 dB 72 dB 72 dB 72 dB p = 0.44 p = 0.03 p = 0.03 p = 0.03 p = 0.03 duration 200 ms 200 ms 50 ms 200 ms 150 ms 150 ms 200 ms standard intensity 85 dB 85 dB 85 dB 79 dB 85 dB 85 dB 72 dB probability p = 0.80 p = 0.20 p = 0.20 p = 0.20 p = 0.80 p = 0.20 p = 0.44 late-deviant 26 Methods Procedure The auditory-visual version of the distraction paradigm was used for the first and the second study, while the auditory-auditory version of the distraction paradigm was carried out for the third study. In the auditory-visual distraction paradigm, auditory stimuli preceded the visual stimuli by 300 ms (onset-to-onset). Stimulus pairs (auditory-visual stimuli) were delivered at a constant rate of one pair every 1.2 s. Participants were instructed to focus on a small fixation cross appearing in the middle of the screen and to press one response button for even numbers and another response button for odd numbers. In the auditory-auditory distraction paradigm, auditory stimuli were delivered at constant stimulusonset-asynchrony (SOA) of 1.3 s. Participants were instructed in this case to press one response button for the short stimuli and another response button for the long stimuli. Both speed and accuracy were emphasized for the task. Participants were also instructed to focus on the task-relevant information and to ignore any task-irrelevant feature (e.g., pitch). Before the experimental session, participants received one practice block without task-irrelevant stimulus. All reached a hit rate level of at least 85%. To avoid tiredness, participants had a short rest period every five minutes or so. Table 4. Number of stimuli, blocks and duration auditory stimuli/block Study I Study II Study III n = 500 n = 300 n = 200 visual stimuli/block n= 500 n= 300 number of blocks 6 6 12 experiment duration 60 min 36 min 52 min 27 All stimuli were delivered in random order, with the only restriction that at least the first four stimuli of each block were standard trials, and that two deviant/novel trials never appeared consecutively. EEG-Recording The experiments were conducted in a dimly lit, electrically and acoustically shielded room. The electroencephalograms (EEG) were continuously digitized at a rate of 500 Hz by SynAmps amplifier (NeuroScan Inc., Herndon, Va., USA) from Ag/AgCl electrodes. Table 5 shows the analogical band-pass filter, the number of recording positions and the elastic cap used for each study. Table 5. Recording settings bandpass filter Study I 0.10 – 100 Hz EEG leads n = 30 cap Electro-Cap International, Eaton, OH (USA) Electro-Cap International, Eaton, OH (USA) Easy Cap FMS, Munich (Germany) additional positions IM1, IM2 Study II Study III 0.05 – 100 Hz 0.05 – 100 Hz n = 30 n = 20 O1, O2 O1, O2 Eighteen electrode positions, in accordance with the 10–20 system, were common for all the studies: Fp1, Fp2, F7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3, Pz, P4, T6 and Oz. Two additional electrodes were placed at left (M1) and right (M2) mastoids for all the studies. Eight more electrodes were used for the first and the second study: TP3 (halfway between T3 and P3), CP1 (halfway between C3 and Pz), FT3 (halfway between F3 and T3), 28 Methods FC1 (halfway between Fz and C3), and the homonymous positions over the right hemisphere (Figure 3). Also, the IN1 (70% of the distance from the left preauricular point to the inion) and IN2 (70% of the distance from the right preauricular point to the inion) lead positions were used for the first study and the O1 and the O2 electrode positions according to the 10-20 system were used for the second and the third study. Figure 3. In blue, the common electrode positions for all the studies. In pink, additional positions for the first and the second study The electro-occulogram (EOG) was recorded with electrodes attached to the canthus and below the left eye. The common reference electrode was attached to the tip of the nose for all the studies. Data analysis Mean response times (RT), hit rate (HR), error rate (ER) and miss rate (MR) were calculated for the standard and deviant sounds in the auditory-auditory distraction paradigm and for the visual stimuli that preceded the standard and deviant/novel sounds in the auditory-visual distraction paradigm. RTs were 29 calculated for hit trials only. Distraction effects caused by deviant/novel trials were analyzed by means of analysis of variance (ANOVA) for repeated measures with type of auditory stimulus as within-subject factors, performed on the mean RT, HR, ER and MR. Post-hoc analysis of the magnitude of the distraction as a function of the type of auditory stimuli was based on non-pooled contrasts. Ttests comparisons between standard and novel stimuli were carried out in the second study for each condition. ERPs were averaged off-line for each auditory stimulus type separately. Artifact rejection was performed to exclude trials with muscular or ocular activity. The method used in the first study was to reject any intervals which exceeded in ±100 V. For the second and third study, standard deviation was calculated within a slide window of 200 ms along all the EEG channels. Intervals exceeding 30 V at any channel for the second study and 50 V in the horizontal or 40 V in the vertical EOG for the third study were not included in the averaging. Table 6. Recording settings total epoch Study I Study II 1300 ms 1100 ms baseline 100 ms 100 ms off-line filter 0.01 – 30 Hz 0.1 – 30 Hz artifact rejection ± 100 V 30 V SD along EEG 50 V SD in the horizontal and 40 V in the vertical EOG Study III 1200 ms 200 ms 1 – 30 Hz Standard-tone trials immediately following deviant or novel-sound trials and, at least, the first four epochs of the beginning (and after the short break in the middle of each block from the second study) were automatically excluded from the averages. 30 Methods Electrophysiological signatures of deviant/novel sound processing were analyzed in the different waveforms obtained by subtracting the standard-trial ERPs from those elicited to the novel-trial ERPs separately for each condition. Mean amplitudes of the three main waveforms of the ERPs elicited during auditory distraction were analyzed (Table 7) by means of ANOVA for repeated measures with type of auditory trial as within-subject factor. Table 7. Latencies of the analyzed components from deviant onset (in parentheses, the relevant electrode to select the largest peak to center the latency window) MMN (Fz) frequency: 100 – 200 ms duration: 150 – 250 ms intensity: 180 – 280 ms N1/MMN (Cz) Study II 110 – 165 ms MMN (Fz) early-deviant: 150 – 200 ms late-deviant: 144 – 294 ms early n-P3a (Fz, Cz, Pz) 200 – 260 ms P3a (Fz, Cz, Pz) early-deviant: 232 – 332 ms late-deviant: 228 – 328 ms late n-P3a (Fz, Cz, Pz) 285 – 345 ms RON (Fz) 460 – 560 ms RON (Fz) early-deviant: 378 – 528 ms late-deviant: 356 – 506 ms Study I Study III When required, scalp distribution analyses were performed on ERP amplitudes at 15 electrodes (F7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3, Pz, P4, T6) after normalization, to prevent genuine differences in scalp distribution from being washed out by amplitude differences. This normalization was done by dividing the amplitude at each electrode by the square root of the sum of the squared amplitudes at the selected electrodes (McCarthy and Wood, 1985). The scalp distribution ANOVA included three 31 factors: frontality (frontal/central/parietal), laterality (five levels from left to right) and stimulus type conditions. In the first study, as the MMN inverts in polarity when the nose is used as the reference electrode, significant activation of MMN generators was analyzed by a t-test comparison between its mean amplitude at Fz and the left mastoid (M1). Comparison of MMN across deviant conditions was carried out on MMN mean amplitudes at Fz. In the second study, mean ERP components and behavioral measures were compared between the headphones and the 0º locations. The remaining positions were analyzed with angle (18º, 72º) and hemifield (right, left) as additional factors in the ANOVA. In the third study, the N2 and the P3b ERPs elicited to the target stimuli were also analyzed. Mean amplitude of the N2 (346-396 ms) and the P3b (496-596 ms) components for the standard targets were compared by t-test analysis. The stimulus duration factor was included in the ANOVAs as an additional factor both for behavior and ERPs analyses. Moreover, mean amplitude to the P3b component for the deviant stimulus (early-DEV: 566666, late-DEV: 632-732 ms) was analyzed by means of an ANOVA for repeated measures with deviant type (early-DEV, late-DEV) and stimulus duration (short, long) as within-subject factors. Peak-latencies over Fz of MMN, P3a and RON were analyzed independent ANOVAs with deviant conditions (early-DEV, late-DEV) and stimulus duration (short, long) as factors. ANOVAs were carried out with the Greenhouse-Geisser correction. When appropriate we reported F value, the uncorrected degrees of freedom, probability level following correction, and the 2 effect size index. 32 Results 4. Results 4.1. Study I Escera, C., Corral, M.J. & Yago, E. (2002). An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, duration and intensity changes. Cognitive Brain Research, 14, 325-332. 33 34 Results 35 36 Results 37 38 Results 39 40 Results 41 4.2. Study II Corral, M.J. & Escera, C. (2008). Effects of sound location on visual task performance and electrophysiological measures of distraction. Neuroreport, in press. Subject: NeuroReport: Decision on Effects of sound location on visual task performance and electrophysiological measures of distraction From: NeuroReport Data: Wed, 16 Jul 2008 13:04:05 -0400 To: cescera@ub.edu 16 July 2008 Dear Professor Escera, Submission no.: NR-D-08-6766R1 Article title: Effects of sound location on visual task performance and electrophysiological measures of distraction Corresponding author: Prof. Carles Escera Your revised manuscript has just been received and checked, and I find the paper now acceptable for publication. I am pleased to let you know that we will therefore publish it in the next issue of NeuroReport to be compiled. Thank you for sending your interesting work to our Journal and for your patience with the revision. Yours sincerely, Giorgio Gabella, MD DSc Editor in Chief NeuroReport 42 Results Effects of sound location on visual task performance and electrophysiological measures of distraction MJ Corral and Carles Escera Cognitive Neuroscience Research Group, Department of Psychiatry and Clinical Psychobiology, Faculty of Psychology, University of Barcelona Correspondence to: Carles Escera, PhD, Professor Department of Psychiatry and Clinical Psychobiology Faculty of Psychology University of Barcelona P. Vall d’Hebron 171 08035-Barcelona, Catalonia-Spain tel.: +34 93 3125048 fax: +34 93 4021584 email: cescera@ub.edu Character count (with spaces): 16,259 43 Abstract Novel sounds embedded in a repetitive stream of auditory stimuli impair performance of the visual task at hand. Parmentier et al., [1] suggested that this distraction effect may be due to the shifting cost of moving attention from the task-irrelevant (auditory) to the task-relevant (visual) channel, or from their shifting of spatial locations. Here, the source location of the sounds in an audio-visual distraction paradigm was varied systematically (headphones and 0, -18, -72, 18, and 72 degrees), and the results revealed significant distracting effects of novel sounds occurring in the headphone and the right location conditions. This supports the assumption that in the behavioral cost observed in the audio-visual distraction paradigm a spatial shift of attention is involved. Keywords: evoked potentials, audition, involuntary attention, orienting response, interlateral asymmetry, hemineglect, pseudo-neglect 44 Results Introduction Research into the neural mechanisms of involuntary auditory attention has experienced considerable progress in recent years thanks to the development of specific “distraction” paradigms [2,3]. In these paradigms, participants classify auditory [4] or visual [4,5] stimuli while ignoring concurrent, task-irrelevant auditory information. The occurrence of any stimulus change in the task-irrelevant auditory channel leads to prolonged response times (RT) in the auditory or visual classification task, reflecting “distraction” of current task performance. This “distracted” performance is accompanied by a pattern of event-related brain potentials (ERPs), the “distraction potential” [3], which includes the mismatch negativity (MMN; [6,7]) and an N1 enhancement for novel sounds [5], the novelty-P3 [8], and a new ERP component discovered in this context of behavioral distraction, the reorienting negativity (RON; [9-11]). The distracting effects in the auditory-auditory distraction paradigm might be accounted by the moving of attention from one auditory dimension (e.g., task-relevant duration) to another (e.g., a taskirrelevant pitch change) of the same perceptual object, leading to large behavioral effects [2-4,11,12]. Less obvious is which stage of visual task-performance is interfered by the occurrence of the task-irrelevant sound change. Alho et al. [13] found that the visual target N1 ERP component was attenuated when the preceding sound deviated in pitch compared to the preceding standard frequency, suggesting that early visual target processing was impaired in distracting trials. However, this could not be confirmed for novel sounds in a study in which visual stimulus identification was made harder by masking the visual targets with a white noise mask [1]. This manipulation enlarged 45 considerably the RT as expected, but left the magnitude of distraction similar to that of a control condition without mask, suggesting that the distracting effect of novel sounds occurred “before” the presentation of the imperative visual stimulus. To confirm this hypothesis, Parmentier and colleagues [1] flashed quickly (50 ms) a shrinking cross in the time period between the auditory, task-irrelevant stimulus and the visual target. They found that the distracting effects of novel sounds were abolished in this condition, suggesting that the RT increase observed in the distracting trials would reflect the time necessary to move the attention drawn to the auditory distracters back to the visual modality upon the occurrence of a fresh sensory event in the task-relevant channel, i.e., vision. However, as discussed by the authors [1], this pattern of results could be also explained by the time necessary to move attention between spatial locations instead of sensory modalities, as the visual stimuli were displayed in a screen located in front of the subject, whereas the sounds were presented through headphones. The present study investigated the role of the spatial source of the auditory stimuli on distraction of visual task-performance. For this purpose, sound location was varied systematically across conditions whereas visual target location remained constant. Notice that we did not intend to assess the effects of distracter location on visual taskperformance [14], but those of varying the location of the sound source on distraction. If distraction depends on moving attention between sensory modalities, having the source of the task-irrelevant auditory stimuli in different spatial locations should not affect the distracting effects of these sounds. However, if a spatial factor is involved, larger distracting effects of novel sounds on visual task-performance, and 46 Results eventually on accompanying ERP signs of distraction, should be obtained with larger spatial separation between the auditory and the visual stimulus sources. Methods Subjects Thirteen healthy students (19-30 years; mean age 23 ys; 6 females) participated in the study, which was approved by the Ethical Committee of University of Barcelona. According to their own report, all subjects were right-handed. As the purpose of the study was to investigate the possible role of sound source location on distractibility, only subjects that showed behavioral distraction in a pilot condition using headphones were included in the sample. Informed consent was obtained from each subject. Stimuli Visual stimuli were the digits 1 to 8 presented equiprobably on a computer screen placed at 1.15 m from the subject’s eyes. The vertical and the horizontal viewing angles were of 2º and 1.25º, respectively. Two types of auditory stimuli were presented with the same duration (150 ms, rise/fall 10 ms): standard and novel. Standard stimuli were pure tones of 600 Hz occurring with a probability of p=0.8. Novel stimuli (p=0.2) were 60 environmental complex sounds selected as the more familiar from a previous study [15]. The auditory stimuli preceded the visual stimuli in 300 ms (onset-to-onset). All stimuli were generated and delivered by means of the Stim module of Neuroscan system (Vernon, VA, USA). 47 Procedure Subjects were instructed to press with their thumbs, as fast and as accurate as possible, one response button for even and another response button for odd numbers, while ignoring the sounds. Response buttons was counterbalanced across subjects. The task was administered in blocks (one per condition) of 300 pairs of stimuli (auditory-visual trials) at a constant inter-trial-asynchrony of 1200 ms. In different source location conditions, the auditory stimuli were delivered from different spatial locations through loudspeakers located, with regard to the subject’s head, at 0º (i.e., by the computer screen), and at -18, -72, 18, 72 degrees (Fig. 1a). In a further condition, the auditory stimuli were delivered through headphones. The sound intensity of all conditions was adjusted to 85 dB SPL at subjects’ ears. Before the experimental session, subjects received one practice block without any auditory stimulation, and all of them reached a hit rate level of at least 90%. Short rest periods were allowed at the middle and the end of each block. EEG-recording The electroencephalogram (EEG) was continuously digitized at a rate of 500 Hz (bandpass 0.05-100 Hz) by SynAmps amplifier (NeuroScan Inc., Herndon, VA., USA) from 30 scalp tin electrodes (Electro-Cap Int., Eaton, OH) positioned at Fp1, Fp2, F7, F3, Fz, F4, F8, T5, T3, C3, Cz, C4, T4, T6, P3, Pz, P4, O1, Oz, and O2, and at ten additional positions (FC1, FT3, CP1, TP3 and LM, and the homologous positions over the right hemisphere). The electro-occulogram (EOG) was recorded with electrodes attached to the canthus and below the left eye. The common reference electrode was attached to the tip of the nose. 48 Results Data analysis Mean RT, hit rate (HR), error rate (ER) and miss rate (MR) were calculated for the digits preceded by standard and novel stimuli separately. Only responses between 100 and 800 ms after visual stimulus onset were included in analyses. RTs were calculated only for hit trials. No button press or a response 100 ms before or 800 ms after visual stimulus onset computed in the MR. Distraction effects caused by novel stimuli were analyzed by means of analysis of variance (ANOVA) for repeated measures with the type of auditory stimulus (standard, novel) and location (headphones, 0º, -18º, -72º, 18º, 72º) as factors. Subsequent analyses were carried out by Student’s t-tests between standard and novel stimuli for each location. The magnitude of the distraction was analyzed, in all possible pairs of locations, by Student’s t-tests on the arithmetical subtraction obtained from the RT to the novel minus the RT to the standard trials. The Bonferroni correction was applied when appropriate. ERPs were averaged off-line for each auditory stimulus type separately over an epoch of 1100 ms, including a pre-auditory stimulus baseline of 100 ms. An artifact rejection algorithm was applied to exclude trials with excessive muscular or ocular activity. This algorithm computed the standard deviation of the amplitude within a slide window of 200 ms along the entire EEG recording, and excludes from averaging any epoch exceeding by 30 V from this standard deviation at any of the electrodes. The first five epochs at the beginning and after the break of each block were excluded also from averaging. After applying these excluding criteria, a minimum of 42 novel and 168 standard trials was included in every condition, there being no significant differences in the number of averaged trials 49 between conditions. Individual ERPs were band-pass filtered between 0.1 and 30 Hz. The electrophysiological signatures of novel sound processing were analyzed on the difference waveforms obtained by subtracting the standard-trial ERPs from the novel-trial ERPs, for each source location condition separately. Mean amplitudes of the three main waveforms of the distraction potential were analyzed in latency windows centered at the corresponding peak in the specified electrodes: N1- enhancement/MMN (Cz, 110-165 ms), early novelty-P3 (200-260 ms) and late (285-345 ms) novelty-P3 (F7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3, Pz, P4, P6), and RON (F3, Fz, F4, 460-560 ms). For each ERP component, the headphones and the 0º source location condition were compared against each other. The remaining source locations were analyzed with angle (18º, 72º) and hemispace (left, right) as factors in the ANOVA. In the analysis of the novelty-P3, the factors phase (early, late), frontality (3 levels) and laterality (5 levels) were also included en the ANOVAs. When appropriate, degrees of freedom were corrected with the Greenhouse-Geisser method, and the reported F, p-values and correction. 2 effect size index were based on this Results Subjects had an overall hit rate of 86%. The ANOVAs revealed no significant effects on HR, ER, and MR. However, visual RT was on average 10 ms slower when the preceding auditory stimulus was a novel sound compared to when it was a standard tone [F(1,12) = 12.02, p = 0.005, 2 = 0.50; Fig. 1b]. The behavioral distraction when 50 Results the sounds were presented through the headphones was of 8 ms (t(12) = -2.44, p = 0.031). -- Insert Figure 1 about here -- Subsequent paired t-test comparisons revealed, after the Bonferroni correction for multiple comparisons, that the effect on RT was due to a significant distracting effect of novel sounds when sounds where presented at the 18º [15 ms; t(12) = -4.44, p = 0.001] and 72º [19 ms; t(12) = -4.47, p = 0.001] sound source location conditions. Moreover, the magnitude of the distracting effect showed a linear increase from left to right, as supported by the significant linear regression between sound source location and the median RT difference between novel and standard trials [F(1,3) = 24.821, p = 0.016; Fig. 1c]. The statistical analyses on the electrophysiological data did not reveal any significant differences on the mean amplitudes of the N1/MMN, novelty-P3 and RON components for the headphones and 0º source location conditions (Fig. 2a). Moreover, no significant differences were found for the N1/MMN and RON components for the remaining source location conditions. The only significant difference was found to the angle factor upon the novelty-P3 [F(1,12) = 5.55, p = 0.036, 2 = 0.50; Fig. 2b], which indicated that novelty-P3 was larger over its two phases when the sounds were presented in source locations closer to the computer screen (-18º and 18º) than when they occurred in the locations far apart from the computer screen (-72º and 72º). -- Insert Figure 2 about here -- 51 Discussion Subject’s performance to classify digits into odd and even categories was affected, overall, by the unexpected occurrence of novel sounds, as shown by slower RTs to visual target stimuli that were preceded by a novel stimulus than to those preceded by a repetitive standard tone. This behavioral distraction is in agreement with many previous studies using similar paradigms with visual and auditory primary tasks (see [3] for an extensive review). However, subsequent analyses revealed significant effects only at the headphones condition, and when the sounds were presented from sources located only on the right hemispace (i.e., 18º and 72º). In other words, novel sounds had no distracting effects on visual task-performance when they occurred by the computer screen, nor when they originated from a sound source located in the left hemispace. The most relevant finding of the present study was that the distraction effect of novel sounds was abolished when the sounds shared the same spatial location as the task-relevant visual stimulation, i.e., in the 0º sound source location condition. This result militates against one of the suggestions of Parmentier el al. [1], proposing that distraction in the auditory-visual distraction paradigm would be due to the time required to move attention from the auditory modality to the visual channel at the occurrence of a fresh sensory stimulus in the taskrelevant channel. In addition, the increase of the distracting effect as a function of the angular distance between the sound source and the target location support the involvement of shifting spatial attention as the cause of the distracting effects involved in the auditory-visual distracting paradigm, as already suggested tentatively by Parmentier et al. [1]. Moreover, the role of the need of shifting spatial attention to 52 Results elicit distraction in the audio-visual distracting paradigm would explain why previous studies using loudspeakers located by the visual targets as the sound source failed to eliciting measurable distracting effects using deviant syllables [16]. A second interesting finding was that the distraction effects of novel sounds were only significant when they occurred in spatial source locations by the right side of the space. In fact, these effects were, according to our regression analysis, larger the farther to the right the sounds were delivered with regard to the computer screen location, supporting the view that auditory spatial attention is distributed in a gradient fashion from peripheral auditory space [17-18]. These results are in agreement with a series of studies on involuntary orienting of attention in the visual modality, showing that pop-out or prime stimuli occurring in the right visual hemifield have larger disturbing effects than those occurring in the left hemifield [19-21]. In addition, they also match with MMN data showing an advantage for processing auditory spatial information in the right compared to the left hemispace [22, 23]. Taking together, our results support the view that the involvement of left and right brain regions in the control of automatic attention on the right hemispace provides a rightward bias for the behavioral influence of unexpected stimuli in right side of the environment [24]. The only remarkable effect on the electrophysiological data was that the novelty-P3 was larger for novel sounds originating from sound source locations near by the computer screen than for remote locations. Several studies have shown the sensitivity of the novelty-P3 to attentional manipulations [5,8,15,25], and one may therefore argue that novel sound occurring closer to the attended extrapersonal space, i.e., by the computer screen, were expected to elicit larger responses 53 than those occurring far away from the attentional set. On the other hand, the lack of relationship between the amplitude of the novety-P3 and the observed behavioral disruption of visual-task performance, give support to recent views that challenge an explicit relationship between the novelty-P3 and the orienting response [3]. Conclusion The results obtained showed that the distracting effects of novel sounds observed in a condition where the sounds were presented through headphones disappeared when the sounds occurred at the same location of the visual targets. Moreover these distracting effects were of increasing magnitude as the angular distance between the sound source and the visual location increased, all in all supporting the role of shifting spatial attention in the well-characterized auditory-visual distraction paradigm. Acknowledgements: The authors would like to thank Sabine Grimm for comments on an early version of the manuscript. This work was supported by grants from the Spanish Ministry of Education and Science (SEJ2006-00496/PSIC; Consolider-Ingenio 2010-CSD-200700012) and the Generalitat de Catalunya (SGR2005-00953). 54 Results References 1. Parmentier FB, Elford G, Escera C, Andres P, San Miguel I. The cognitive locus of distraction by acoustic novelty in the cross-modal oddball task. Cognition 2008; 106: 408-432. 2. Escera C, Alho K, Schroger E, Winkler I. Involuntary attention and distractibility as evaluated with event-related brain potentials. Audiol Neurootol 2000; 5: 151-166. 3. Escera C, Corral MJ. Role of mismatch negativity and novelty-P3 in involuntary auditory attention. Int J Psychophysiol 2007; 21: 251-264. 4. Schroger E, Wolff C. Behavioral and electrophysiological effects of task-irrelevant sound change: a new distraction paradigm. Brain Res Cogn Brain Res 1998; 7: 71-87. 5. Escera C, Alho K, Winkler I, Naatanen R. Neural mechanisms of involuntary attention to acoustic novelty and change. J Cogn Neurosci 1998; 10: 590-604. 6. Naatanen R, Michie PT. Early selective-attention effects on the evoked potential: a critical review and reinterpretation. Biol Psychol 1979; 8: 81-136. 7. Escera C. The mismatch negativity 30 years later: how far have we come? Int J Psychophysiol 2007; 21: 129-132. 8. Friedman D, Cycowicz YM, Gaeta H. The novelty P3: an eventrelated brain potential (ERP) sign of the brain's evaluation of novelty. Neurosci Biobehav Rev 2001; 25: 355-373. 9. Horvath J, Maess B, Berti S, Schröger E. Primary motor area contribution to attentional reorienting after distraction. Neuroreport 2008; 19: 443-446. 10. Escera C, Yago E, Alho K. Electrical responses reveal the temporal dynamics of brain events during involuntary attention switching. Eur J Neurosci 2001; 14: 877-883. 55 11. Schroger E, Wolff C. Attentional orienting and reorienting is indicated by human event-related brain potentials. Neuroreport 1998; 9: 3355-3358. 12. Berti S, Roeber U, Schroger E. Bottom-up influences on working memory: behavioral and electrophysiological distraction varies with distractor strength. Exp Psychol 2004; 51: 249-257. 13. Alho K, Escera C, Diaz R, Yago E, Serra JM. Effects of involuntary auditory attention on visual task performance and brain activity. Neuroreport 1997; 8: 3233-3237. 14. Roeber U, Widmann A, Schroger E. Auditory distraction by duration and location deviants: a behavioral and event-related potential study. Brain Res Cogn Brain Res 2003; 17: 347-357. 15. Escera C, Yago E, Corral MJ, Corbera S, Nunez MI. Attention capture by auditory significant stimuli: semantic analysis follows attention switching. Eur J Neurosci 2003; 18: 2408-2412. 16. Schirmer A, Li Q. Attention capture by change in speaker emotion differs between men and women. Soc Cogn Affect Neurosci (in press). 17. Munte TF, Kohlmetz C, Nager W, Altenmuller E. Neuroperception. Superior auditory spatial tuning in conductors. Nature 2001; 409: 580. 18. Nager W, Kohlmetz C, Altenmuller E, Rodriguez-Fornells A, Munte TF. The fate of sounds in conductors' brains: an ERP study. Cogn Brain Res 2003; 17: 83-93. 19. Castro-Barros BA, Righi LL, Grechi G, Ribeiro-do-Valle LE. Interlateral asymmetry in the time course of the effect of a peripheral prime stimulus. Brain Cogn 2008; 66: 265-279. 20. Pollmann S. A pop-out induced extinction-like phenomenon in neurologically intact subjects. Neuropsychologia 1996; 34: 413-425. 21. Pollmann S. Extinction-like effects in normals: independence of localization and response selection. Brain Cogn 2000; 44: 324-341. 56 Results 22. Kaiser J, Lutzenberger W, Birbaumer N. Simultaneous bilateral mismatch response to right- but not leftward sound lateralization. Neuroreport 2000; 11: 2889-2892. 23. Dimitrijevic A, Stapells DR. Human electrophysiological examination of buildup of the precedence effect. Neuroreport 2006; 17: 1133-1137. 24. Mesulam MM. Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Ann Neurol 1990; 28: 597-613. 25. SanMiguel I, Corral MJ, Escera C. When loading working memory reduces distraction: behavioral and electrophysiological evidence from an auditory-visual distraction paradigm. J Cogn Neurosci 2008; 20: 1131-1145. 57 Figure 1. a) Outline of the experimental setting. Subjects sat 115 cm in front of a computer screen, and the sounds were presented, in different source location conditions, from headphones (H) or from loudspeakers located at 72º and 18º degrees of either side, and from a central location, i.e., by the computer screen (0º). b) Response time (RT) for standard (std) and novel (nov) trials in each location condition. c) Behavioral distraction(novel RT minus standard RT) as a function of location. * p < 0.05, ** p < 0.01, *** p < 0.001 58 Results Figure 2. a) Difference waveforms obtained by subtracting the novel minus de standard ERPs at Fz and Cz. b) Scalp distribution maps of the two phases of the novelty-P3 (early: 200-260 ms; late: 285-345 ms) for all source location conditions. 59 4.3. Study III Corral, M.J., Berti, S., Jacobsen, T., Widmann, A., Yago, E., Schröger, E., & Escera, C. (submitted). Distraction effects with different distractor-totarget intervals: a combined behavioral and event-related brain potential study. 60 Results Distraction effects with different distractor-to-target intervals: a combined behavioral and event-related brain potential study MJ Corral1, Stefan Berti2, Thomas Jacobsen3, Andreas Widmann3, Elena Yago1, Erich Schröger3 and Carles Escera1,* 1 Cognitive Neuroscience Research Group, Department of Psychiatry and Clinical Psychobiology, Faculty of Psychology, Universitat de Barcelona. Passeig Vall d’Hebron 171, 08035-Barcelona, CataloniaSpain. 2 Psychologisches Institut, Johannes Gutenberg-Universität Mainz, Staudinger Weg 9, D-55099 Mainz, Germany. 3 Institut für Psychologie I, Universität Leipzig, Seeburgstr. 14-20, D04103 Leipzig, Germany. * Corresponding author Number of text pages (including figures and tables): 20 Number of figures and tables: 4 Address correspondence to: Carles Escera, PhD, Professor Cognitive Neuroscience Research Group. Department of Psychiatry and Clinical Psychobiology. Faculty of Psychology, University of Barcelona. P. Vall d'Hebron 171, 08035 Barcelona. Catalonia-Spain Tel: +34-933-125-048 Fax: +34-934-021-584 email: cescera@ub.edu URL address: www.ub.edu/brainlab 61 Abstract In order to evaluate whether the temporal distance between the taskirrelevant change and the task-relevant information is critical for distraction and whether the RON component indicate a pure signal of reorientation to the task independently to deviance onset, slight pitch changes were introduced in an auditory duration discrimination task while the target onset remained constant. Impaired performance was observed after each deviance onset. However, the pattern of distraction was different as a function of the deviance onset (at 50 or 150 ms) and the type of target processing (short or long stimuli). The RON component was not strongly synchronized to the target onset. The present results contribute to the mental chronometry of auditory distraction by showing that task-irrelevant information affects different stages of target processing and the reorienting of attention to the task is not totally independence of the distractor processing. Descriptors: human, audition, mismatch negativity (MMN), P3a, reorienting negativity (RON), deviant asynchrony, impulsive style, reflexive style, slow responses, inaccurate responses. 62 Results Introduction In everyday live, our senses are constantly stimulated by information. The adaptive control of behavior requires the integration and coordination of relevant information for our task at hand and the flexibility to react to unexpected new events. Driving a bike requires focus on the way and being able to detect unexpected changes around, for instance the sudden appearing of a child. The ability to respond to changes is maintained by the capability of the sensory systems to detect irregularities without focused attention which, subsequently, may result in the allocation of attention onto new information. However, it is still unclear under what circumstances a switch of attention after the detection of a change is accomplished. In a series of studies the potential of physical changes to catch attention was tested by so called distraction paradigms demonstrating response times prolongation on a primary task by changes of a task irrelevant stimulus feature (Berti & Schroger, 2003; Escera, Alho, Winkler, & Naatanen, 1998; Escera, Corral, & Yago, 2002; Escera, Yago, Corral, Corbera, & Nuñez, 2003; Schroger & Wolff, 1998a,b; Schroger, Giard, & Wolff, 2000). The advantage of this distraction logic is that the processing of a non-attended physical information could be tested. These studies established that the behavioral distraction effects depend on the size (Berti, Roeber, & Schroger, 2004), the unpredictability (Munka & Berti, 2006; Sussman, Winkler, & Schroger, 2003), and the channel proximity of the change to the task-relevant information (Schroger & Wolff, 1998a). That is, large and unpredictable distractors such as a telephone ring may obviously yield larger distraction than small distractors, which are physically similar to standard auditory background, or than predictable stimuli (e.g., visually 63 cued ones). On the other hand, there is no systematic investigation how a distractor interferes to cause impaired performance. In detail, the question is whether the task-irrelevant stimulus may affect to any stage of the task-relevant processing or, whether there is a particular vulnerable temporal window between the distractor and the target stimulus which determines more distraction. In the present study we investigated this question by manipulating the onset of the deviant stimulus information without changing other aspects of the stimuli and the task itself. In a previous study, Schröger (1996) showed impoverished performance (hit rate decrease and response time increase) when the task-irrelevant stimulus change was presented at 200 ms from the task-relevant tones, whereas no detrimental performance was found when the interval was enlarged to 560 ms. Later, Escera, Yago, & Alho, 2001 also used two different asynchronies between the taskirrelevant sound and the subsequent imperative visual stimulus. Participants were slower to classify the visual stimulus when it was preceded by a slightly higher deviant tone or a novel environmental sound than when the visual target was preceded by a repetitive standard tone. However, no performance differences were found irrespective of whether the asynchrony between the task-irrelevant sound and the task-relevant visual stimuli was of 245 or 355 ms. Moreover, a negative component related to reorienting of attention to towards main task performance after distraction (RON) appeared synchronized to the task-relevant information. Different temporal distances between distractor and target onset has been manipulated in an auditory duration discrimination task, where distractor onset coincided with the stimulus onset and the temporal 64 Results difference between the short and the long stimuli determines target onset. For example, when participants are to classify short stimulus of 200 ms and long stimulus of 400 ms, the distractor (task-irrelevant change) begins with stimulus onset, whereas the task-relevant target onset occurs 200 ms later (when the stimulus should finished, if it was a short stimulus). With this paradigm, it has been shown that asynchronies between deviance onset and target onset of 200 ms (Roeber, Berti, & Schoger, 2003; Roeber, Widmann et al., 2003; Schroger & Wolff, 1998b; Schroger et al., 2000) and 100 ms (Schroger & Wolff, 1998a) are effective enough to cause behavioral distraction. Moreover, Rinne, Sarkka, Degerman, Schroger, & Alho (2006) showed that task-irrelevant decrements or increments of intensity when participants had to discriminate between two equiprobable sounds which differed in pitch, that is, when the distance between the taskrelevant and the task-irrelevant feature was of 0 ms, also caused distraction. In addition, (Roeber, Widmann et al., 2003) found slower responses when infrequent long stimuli (400 ms) appeared in a sequence of repeated short stimuli (200 ms) and participants had to discriminate the localization of the sound (front or left), that is, when the task-irrelevant feature came 200 ms after the task-relevant information. So it seems that task-irrelevant changes presented from 355 ms to 200 ms relative to the target onset could impair performance. The present study addressed two specific questions. First, whether the occurrences of the auditory change at different asynchronies from the target feature onset could affect the magnitude of distraction. Second, whether the reorienting of attention indexed by RON is insensitive to the deviance onset, as in the previous study of Escera, Yago, & Alho, 2001. For that purpose, participants were instructed to respond to 65 short (200 ms) and long stimulus (400 ms) and ignore all other aspects of stimulation. Most of the tones (p=0.88) had the same pitch from the beginning to the end (standard tones, STD) and in a small set of stimuli (p=0.12) slight task-irrelevant pitch changes were introduced (deviant stimuli, DEV). Two time points (fig. 1) were selected maintaining a similar distance between them around 100 ms as the previous study (Escera, Yago, & Alho, 2001). One pitch deviance was presented early, at 50 ms from stimulus onset (early-DEV). The other pitch deviance was presented later, at 150 ms from stimulus onset (late-DEV). Because the short stimuli were of 200 ms and the long stimuli of 400 ms, the discrimination point between them occurred at 200 ms from stimulus onset (point when the short stimulus finished and the long stimulus continued). So, target onset remained constant through deviant stimuli. Whether a temporal window that determines more distraction exists, different behavioral responses should be recorded between conditions. Moreover, whether RON shows the reorienting of attention to the relevant information, it should be peaked at the same time window between deviant conditions, because for both deviant conditions target onset began at the same time. Methods Participants Twelve healthy students of the University of Leipzig (18-25 years; mean age 21 ys; 2 males) participated in the study for either course credit or payment (12 €). All of them were right-handed and none reported auditory dysfunction. One further participant and the long late-DEV ERP condition of another participant were discarded due to 66 Results technical problems. Informed consent was obtained from each participant, after the nature of the study was explained to them. Stimuli Short (200 ms) and long (400 ms) stimuli were presented with equal probability of appearance (p=0.5). The pitch of the stimuli was the same, from the beginning to its end, for standard stimuli (p=0.88), or changed, at some point from sound onset, for the deviant stimuli (p=0.12). That is, standard stimuli (STD) were short or long sinusoidal tones of 1000 Hz, with equal probability of appearance (p=0.44 each one). In turn, deviant stimuli had a first segment of the same pitch as the standard stimuli (1000 Hz), and a second segment of slightly higher pitch (1100 Hz). The transition between the standard and the deviant pitch was carried out over a window of 20 ms centered at the point of change. As shown in Fig. 1, the duration of the initial segment was of 50 ms or 150 ms from stimulus onset. That is to say, for the early deviant stimulus (early-DEV), there was 150 ms of pitch change before target onset (i.e., discrimination point between short and long stimulus 200 ms from stimulus onset), while for the late deviant stimulus (late-DEV), there was 50 ms of pitch change before target onset. As the standard stimuli, both type of the deviant stimuli could be short or long in duration (i.e., 200 or 400 ms), and had equal probability of appearance (p=0.03 each one). All stimuli were delivered binaurally through headphones at intensity of 75 dB SPL and with a rise and fall time of 5 ms. Procedure Participants were instructed to press, as fast and as accurately as possible, one response button for the short stimuli and another response button for the long stimuli. Since only the stimulus duration 67 information was relevant for the task, participants were also instructed to concentrate on stimulus duration and to ignore all other aspects of stimulation (e.g., pitch). The task was administered in 12 blocks of 200 auditory stimuli delivered each, at a constant stimulus-onsetasynchrony (SOA) of 1300 ms. Standard and the two types of deviant stimuli were presented within the same block in random order, with the only restriction that at least the first four stimuli of each block were standard stimuli, and that two deviant stimuli never appeared consecutively. Before the experimental session, participants received one practice block with only standard tones, and all of them reached a hit rate level of at least 85%. To avoid tiredness, participants had a short rest period after each block. In order to reduce eye-blinks and movements during the EEG recording, participants were instructed to focus on a central fixation point. EEG-recording The experiment was conducted in a dimly lit, electrically and acoustically shielded room. The electroencephalogram (EEG) was continuously digitized at a rate of 500 Hz (bandpass 0.05-100 Hz) by SynAmps amplifier (NeuroScan Inc., Herndon, Va., USA) from 20 scalp Ag/AgCl electrodes positioned according to the 10-20 system (Fp1, Fp2, F7, F3, Fz, F4, F8, T5, T3, C3, Cz, C4, T4, T6, P3, Pz, P4, O1, Oz, and O2). The electrodes were mounted in an elastic cap (Easy Cap FMS, Munich, Germany). Two additional electrodes were placed on left (M1) and right (M2) mastoids. The electro-occulogram (EOG) was recorded with electrodes attached to the left canthus and below the left eye. The reference electrode was attached to the tip of the nose. 68 Results Data analysis Mean response times (RT), hit rate (HR), error rate (ER) and miss rate (MR) were calculated for the standard and the two types of deviant sounds, as well as for short and long stimuli, separately. Only responses between 300 and 1200 ms after stimulus onset were included in the computation of RT, HR and ER. RTs were calculated for hit trials only, with respect to the onset of the duration difference between the short and long stimulus, i.e., taking the zero time at 200 ms from stimulus onset. No button press or a response before 300 ms from stimulus onset computed in the MR. Distraction effects caused by deviant stimuli were analyzed by means of analysis of variance (ANOVA) for repeated measures with type of auditory stimulus (STD, early-DEV, and late-DEV) and stimulus duration (short, long) as withinsubject factors, performed on the mean RT, HR, ER and MR. Post-hoc analysis of the magnitude of the distraction as a function of the type of deviant stimuli was based on non-pooled contrasts. ERPs were averaged off-line for each auditory stimulus type separately over an epoch of 1000 ms which included a pre-auditory stimulus baseline of 200 ms. Artifact rejection was performed to exclude trials with extreme EOG activity. Standard deviation was calculated within a slide window of 200 ms along all the EEG. Intervals exceed 50 V in the horizontal and 40 V in the vertical EOG from the standard deviation, as well as, the first four epochs of each block, were automatically excluded from averaging. Individual ERPs were band-pass filtered between 1 and 30 Hz. Target (N2 and P3b) and difference wave (MMN, P3a and RON) ERP components were analyzed. Mean amplitude to N2 and P3b for the short and long standard stimulus was compared by t-test analysis. 69 Mean amplitude to the P3b component for the deviant stimulus was analyzed by means of an ANOVA for repeated measures with deviant type (early-DEV, late-DEV) and stimulus duration (short, long) as within-subject factors. Moreover, P3b scalp distribution was analyzed on the ERP-normalized amplitudes (McCarthy & Wood, 1985) at F7, F3, Fz, F4, F8, T3, C3, Cz, C4, T4, T5, P3, Pz, P4 and T6, by an ANOVA with target type (short STD, long STD, short early-DEV, long early-DEV, short late-DEV, long late-DEV), frontality (frontal, central, parietal) and laterality (left, central, rigth) as factors. Difference waveforms were calculated by subtracting ERPs elicited by standard stimuli from those elicited by deviant stimuli of the same duration. Mean amplitude of MMN at Fz was analyzed by means of an ANOVA with deviant type (early-DEV, late-DEV) and stimulus duration (short, long) as factors. Scalp distribution of P3a and RON was analyzed by an ANOVA with deviance onset (early-DEV, late-DEV), stimulus duration (short, long), frontality (frontal, central, parietal) and laterality (left, central, rigth) as factors. Peak-latencies over Fz of MMN, P3a and RON were analyzed independent ANOVAs with deviance onset (early-DEV, late-DEV), stimulus duration (short, long) as factors. The temporal window for all the components analyzed (in parenthesis, the relevant electrode to select on visual inspection the largest peak to center the latency window) was taken from the stimulus onset to the individual target ERP components (N2 and P3b), and from the deviance onset to the difference waveform components (MMN, P3a and RON). The interval analyzed for N2 (Cz) was 345í395 ms for all the targets. P3b (Pz) was analyzed in the following temporal windows: 495í595 ms for the standard, 565í665 ms for the early-DEV, and 630í730 ms for the late-DEV stimuli. MMN (Fz), P3a (Fz, Cz, Pz) and RON (Fz, Cz, Pz) components were analyzed for the early-DEV 70 Results condition at 150í200 ms, 230í330 ms, 380í530 ms, respectively, and for the late-DEV condition at 145í295 ms, 230í330 ms, 355í505 ms, respectively. ANOVAs were carried out with the Greenhouse-Geisser correction. When appropriated, we reported the F value, the uncorrected degrees of freedom, probability level following correction and the 2 effect size index. Results Performance Across stimulus duration and stimulus types, participants had an overall high performance level of about 92% classifying short and long stimuli. The stimulus duration factor (short, long) did not reach significant differences in hit rate (HR) [F(1,11) = 4.43, p = 0.059, 0.29] or error rate (ER) [F(1,11) = 4.30, p = 0.062, 2 2 2 = = 0.28], although response time (RT) was clearly prolonged for the long than for the short stimuli [F(1,11) = 90.92, p < 0.001, = 0.89] (Fig. 2). Miss responses did not yield significant differences for any factor or interaction. Stimulus type factor (STD, early-DEV, late-DEV) was significant for HR [F(2,22) = 17.39, p = 0.001, 17.45, p = 0.001, 0.001, = 0.83, 2 = 0.58, 2 = 0.61], ER [F(2,22) = = 0.56, 2 = 0.61], and RT [F(2,22) = 61.93, p < = 0.85]. Post-hoc comparisons based on non- pooled contrast revealed that participants were more inaccurate and slower for the deviant than for the standard stimuli [stimulus type (STD, early-DEV): HR, F(1,11) = 18.71, p = 0.001, F(1,11) = 18.46, p = 0.001, 2 2 2 = 0.63; ER, = 0.63; RT, F(1,11) = 48.00, p < 0.001, = 0.81; stimulus type (STD, late-DEV): HR, F(1,11) = 14.73, p = 71 0.003, 2 = 0.57; ER, F(1,11) = 15.18, p = 0.002, 2 2 = 0.58; RT, F(1,11) = 101.93, p < 0.001, = 0.90]. Moreover, post-hoc comparisons 2 between deviant stimuli yielded a distraction trade-off effect. Participants were faster [RT, F(1,11) = 13.92, p = 0.003; more inaccurate [HR, F(1,11) = 15.73, p = 0.002, F(1,11) = 16.16, p < 0.002, late-DEV stimuli. 2 2 = 0.56] but = 0.59; ER, = 0.60] for the early-DEV than for the No significant interactions between stimulus duration and stimulus type were observed in HR or ER, though a significant interaction was found for RT [F(2,22) = 17.07, p < 0.001, = 0.92, 2 = 0.61]. T-test comparisons revealed that participants were slower for the long than for the short standard [t(11) = -4.57, p = 0.001], the early-DEV [t(11) = -5.88, p < 0.001] and the late-DEV [t(11) = -9.32, p < 0.001] stimuli. A specific analysis for the short stimuli did not show significant differences to the stimulus type factor in HR or ER. However, there was a significant difference for the stimulus type factor in RT [F(2,22) = 10.69, p = 0.001, = 0.86, 2 = 0.49]. T-test comparisons revealed that participants were slower for the short deviant stimuli than for the short standard tones [STD, early-DEV: t(11) = -3.12, p = 0.010; STD, lateDEV: t(11) = -5.38, p < 0.001], but not between short deviant stimuli. On the other hand, the same specific analysis for the long stimuli showed significant differences for the stimulus type factor in HR [F(2,22) = 8.67, p = 0.012, 0.010, 0.77, 2 2 = 0.53, = 0.44], ER [F(2,22) = 9.25, p = = = 0.53, 2 = 0.46], and RT [F(2,22) = 66.28, p < 0.001, = 0.86] and whatever it was the t-test comparison (t values rage from -9.95 to 3.25, all p ” 0.025). 72 Results Event-related brain potentials Target ERPs Fig. 3 shows the grand-average ERPs for all the targets. No significant differences between the short and the long standard tones were found for the N2 mean amplitude at Cz or Fz. P3b mean amplitude at Pz was larger for the short than for the long standard tone [t(10) = 2.96, p = 0.014]. Because deviant stimuli were also targets, P3b component were elicited in these trials. An ANOVA with deviant type (early-DEV, late-DEV) and stimulus duration (short, long) as factors did not reveal significant differences over the P3b mean amplitude at Pz for any factor or interaction. As showed in the right part of Fig. 3, similar scalp distributions were found to P3b for all stimulus targets except for the short standard one. An ANOVA with target type (short STD, long STD, short early-DEV, long early-DEV, short late-DEV, long late-DEV), frontality (frontal, central, parietal) and laterality (5 levels) as factors over the normalized P3b mean amplitudes yielded a significant interaction between the target type x frontality x laterality [F(40,400) = 3.18, p = 0.011, 0.12, 2 = = 0.24]. Post-hoc comparison between the short STD P3b and the rest of the targets yielded a significant target type factor (F values raging 5.27 and 9.47, all p ” 0.045). An ANOVA only for frontal leads (F7, F3, Fz, F4, F8) also revealed a significant difference to target type factor [F(5,50) = 6.63, p = 0.002, = 0.58, 2 = 0.40]. Post-hoc comparisons between the short STD P3b and the rest of the targets again yielded a significant effect of the target type factor (F values raging 7.67 and 18.50, all p ” 0.020), which indicated lager central positivity for the short standard than for the remaining targets. 73 Difference waveforms Both deviance onsets elicited a similar pattern of deflections peaked at different latencies (Fig. 4). The first negative deflection showed one small first peak, followed by a larger negativity. The first peak was elicited about 120 ms from the change onsets, so probably indicated the N1 to the pitch change from 1000 to 1100 Hz. The second peak was generated around 175 ms from each deviance onset, so presumable involve the change detector mechanism indexed by MMN. Once corrected the temporal distance between deviance onsets (i.e., 100 ms), an ANOVA with deviance onset (early-DEV, late-DEV) and stimulus duration (short, long) as factors over the peak latencies did not yield statistically significant differences for the MMN component. However, the deviance onset factor was significant for the latencies of P3a [F(1,10) = 16.79, p = 0.002, p = 0.006, 2 2 = 0.63] and RON [F(1,10) = 12.33, = 0.55] components, indicating that P3a and RON components emerged 30 ms and 40 ms, respectively, earlier for the late-DEV condition than for the early-DEV condition. Also, the interaction deviance onset x stimulus duration was significant for the RON component [F(1,10) = 8.62, p = 0.015, 2 = 0.46]. The same analysis for the RON peak-latency from target onset (without deviance temporal correction) yielded analogous statistical results. T-test comparison showed that short and long early-DEV conditions differed significantly [t(11) = -3.54, p = 0.005], but not the short and the long late-DEV nor the short early-DEV and the short late-DEV conditions. Although, MMN appeared at different time windows from the target onset (30 ms for the early-DEV condition and 125 ms for the late-DEV condition), an ANOVA over its mean amplitude with deviance onset (early-DEV, late-DEV), stimulus duration (short, long) and frontal 74 Results electrodes (F7, F3, Fz, F4, F8) did no yield significant difference for any factor or interaction. P3a normalized-mean amplitude analyzed by means of an ANOVA with deviance onset, stimulus duration, frontality and laterality as factors yielded significant differences for the deviance onset factor [F(1,10) = 9.06, p = 0.013, 2 = 0.47] and the deviance onset x 2 stimulus duration interaction [F(1,10) = 6.53, p = 0.029, = 0.39]. Analogous analysis only for early-DEV conditions did not revealed significant differences or interactions, while late-DEV conditions yielded significant difference between the short and the long deviance onset [F(1,10) = 6.70, p = 0.027, 2 = 0.40]. The same analysis only for short deviance onsets showed a significant interaction between deviance onset x laterality factors [F(4,44) = 4.72, p = 0.032, 2 = 0.38, 2 = 0.30], whereas long deviance onsets yielded a significant = difference for deviance onset factor [F(1,10) = 9.42, p = 0.012, 0.48]. The statistical analysis of RON by means of a four-way repeated measures ANOVA yielded significant effects of the deviance onset factor [F(1,10) = 6.87, p = 0.026, 2 = 0.41] and deviance onset x 2 stimulus duration [F(1,10) = 7.18, p = 0.023, duration x laterality [F(4,40) = 3.70, p = 0.031, 2 = 0.42], stimulus 2 = 0.63, = 0.27] and = deviance onset x frontality x laterality [F(8,80) = 4.55, p = 0.007, 0.41, = 0.31] interactions. Analysis only for the early-DEV condition 2 showed the interaction stimulus duration x laterality significant [F(4,44) = 3.40, p = 0.028, = 0.77, = 0.24], while short and long late-DEV conditions did not reveal significant differences or interactions for the main factors. The same analysis only for short deviance onsets showed a significant interaction between deviance onset x frontality x 75 laterality factors [F(8,88) = 4.62, p = 0.012, = 0.33, 2 2 = 0.30], whereas long deviance onsets yielded a significant difference for deviance onset factor [F(1,10) = 8.41, p = 0.016, 2 = 0.46] and the = deviance onset x frontality interaction [F(1,10) = 5.54, p = 0.029, 0.43, = 0.36]. Discussion The main question of the present study was whether the onset of an unexpected (and irrelevant) change of the stimulation may disrupt the processing of task-relevant information. In addition to other studies, we embedded this change within the task-relevant stimulus at different points in time relative to the onset of the relevant or target information to test the effect of the deviant-to-target interval on distraction. In general, all participants were able to perform the task with high efficiency, as indicated by the overall hit rate of above 90%. More important, the RT data demonstrated that the processing of taskrelevant information is vulnerable irrespective of the deviant-to-target interval. This is in line with different other studies demonstrating distractibility within a time-window of -355 to 200 ms relative to target onset. In addition with theses findings our study shows that distractibility is also possible during the processing of task relevant stimulus. This is supported by the ERP results showing the elicitation of distraction potentials, namely MMN, P3a, and RON in both types of deviant stimuli. However, beside this clear cut result of general distractibility in every phase of task-related information there are remarkable differences not only in the processing of the deviant information but also in the processing of the target information. On one hand, all parameters of distraction (RT prolongation, P3a and RON) 76 Results show remarkable differences between deviant stimuli with an early and a late pitch change. On the other hand, even processing of standard stimuli differs between short and long stimuli. These different effects of the manipulation of deviant and target information interact in the parameters of distraction. We, therefore, start with the discussion of the effects of the target information. In general, participants were slower in pressing the response button for the long stimuli than for the short stimuli. Obviously, the short stimuli have a sharp offset that could serve as a cue in order to respond. Mean RT from stimulus offset was on average about 328 ms for the short and 157 ms for the long stimulus. Because 157 ms does not seem enough to prepare and execute a response, the long stimulus response should be prepared some time before. One possible explanation for the delay (30 ms) in responding to the long in comparison to the short stimuli could be due to the time needed by participants to realize that the long stimuli was exceeding the critical duration of 200 ms. Furthermore, in order to resolve the task it could be just necessary activate an internal template of the short duration. This explanation is supported by the different scalp distribution of P3b found for the short standard tone in comparison to the rest of the targets. If P3b reflects the match between stimulus and voluntary maintained attentional trace (Naatanen, 1992), it could interpret that participants held in mind the “short standard template” in order to discriminate short and long stimuli. When the presented target was a short standard tone, it matched with the neural trace saved in working memory and the associated response was triggered faster. However, whether the ongoing stimulus did not match with the neural trace saved, then the prepared response had been stopped and a switch response for the long stimuli was required, resulting in a delay 77 response time. Actually, frontal leads showed significant different activity for all the targets that were not short standard tones, that is, for long and deviant stimuli. This topographic difference could be interpreted as a result of a context update triggered by all the targets that they were not match with the short standard tone memory trace. Moreover, the occurrence of a task-irrelevant pitch change in the short deviant stimuli did not cause hit rate decrease or error rate increase. Participants classified without difference the short standard and the short deviant stimuli. If we assume that HR and ER are measures of stimulus classification processing, it seems clear that unexpected pitch changes were not able to affect this target processing stage. The switch of attention indexed by P3a peaked from the target onset was around 150 ms for the early-DEV condition and around 225 ms for the late-DEV condition. It could be possible for the early-DEV condition the classification was still not ready, while for the late-DEV condition was advanced or already made. This could explain the underlying slight lateralized activity for the late-DEV than the early-DEV condition in the P3a window. However, although HR or ER was not affected by the task-irrelevant changes, RT was slower for the short deviant than for the short standard stimuli. Specifically, the delay was 40 ms which coincided with the temporal window that covered P3a, so it could be possible P3a generation was the cause of the delay and responses were postponed until its end. A different pattern of results was obtained for the long stimuli. The occurrence of a task-irrelevant pitch change caused hit rate decreased, error rate increase and slower response time to the long deviant in comparison to the long standard and the short deviant stimuli. Maybe because of the long stimulus processing was delayed 78 Results in comparison to the standard stimuli, or because a switch response should be produced, or both, an inaccurate and slower performance was registered. Interesting, there was a trade-off between response time and response accuracy of pitch change effects in the two long deviant stimuli. The long early-DEV stimulus resulted in more inaccurate responses (decreased hit rate caused by an increased error rate) than when the change occurred closer to discrimination point, i.e., the late-DEV stimulus. Whereas, the response time increase was smaller for the long late-DEV stimulus than for the long early-DEV stimulus. In this way, it appears that a task-irrelevant pitch change occurring 150 ms before target onset yielded an impaired impulsive response (fast but inaccurate), while the same taskirrelevant change occurring 100 ms later, i.e., 50 ms before target onset, resulted in a impaired reflective response (slow but accurate). Again, a look at the underlying neuroelectric brain activity may help to elucidate these results. The P3a scalp distributions for long stimuli clearly showed that the underlying processing was significant different between deviance onsets. Short and long early-DEV conditions did not show significant differences, which it could mean that similar switches of attention was triggered to the temporal window of 150 ms from the target onset. Actually, the statistical analyses did not reveal significant interaction between stimulus type and stimulus duration in HR or ER, which it meant that similar stimulus classification was obtained between short and long early-DEV stimulus. However, the scalp distribution of P3a to the long late-DEV condition clearly differed from the rest of conditions. The P3a component for the late-DEV condition peaked around 225 from the target onset, which it could be explain that the classification process was presumably quite advance yielding better rate that the long early-DEV stimulus. Moreover, although P3a 79 peaked at the same temporal window to the short and the long lateDEV conditions, presumably the underlying stimulus processing should be delay for the long stimulus in comparison to the short stimulus. It could explain the different scalp distributions found between the short and the long late-DEV conditions. Also, the slower RT to the long late-DEV than the short late-DEV stimuli could due that the P3a affected different stages of the response processing, a latter in the case of the short and earlier in the cause of the long stimuli. Therefore, the present results suggest that distraction it is not only observed when the processing of the task-relevant object just started (i.e., early-DEV stimulus), but also when it has been processed even during 150 ms (i.e., late-DEV stimulus), all in all suggesting that the different auditory features are processed and controlled independently from each other, and supporting the hypothesis of feature specific distraction (Jankowiak & Berti, 2007). In this line, one may argue that task-irrelevant and task-relevant features are processed in parallel until the orienting of attention, indexed by P3a, is triggered. At this point of time, the behavior effects will depend on the ongoing processing. This explanation could also help to elucidate the results obtained by (Roeber, Widmann et al., 2003). In their data, when the distractor preceded the target onset for 200 ms, P3a peaked before response time and clear distraction was observed. However, when the distractor appeared 200 ms after the target onset, P3a peaked after the response time and no impaired performance was observed. Also, the hypothesis of auditory independent features fits well with the different behavioral patterns produced by novel stimuli observed in several previous studies. Opposite to what occurs with deviant sounds, novel sounds do not usually lead to hit rate decrements but to large response time increases (Escera et al., 1998; Escera, Yago, & 80 Results Alho, 2001). As Parmentier et al. (2008) have shown in the auditoryvisual paradigm, the slower response times to the visual target preceded by a novel sound, in comparison to a standard tone, are not due to a detrimental visual analysis nor categorical processing of the target stimuli, but just to the time involved in recapturing attention from auditory novel processing to visual target processing. It seems that novel task-irrelevant sounds produce a complete reallocation of attentional resources and once the relevance of the novelty has been evaluated, attention could return back to target processing, which results in preserved hit rate but delayed response times. On the other hand, deviant stimuli allow parallel processing of the task-relevant and task-irrelevant information affecting different behavioral responses that reflect impaired stages from the target processing. On the contrary to a previous study (Escera, Yago, & Alho, 2001), RON component did not appear completely synchronized to the target onset. One possible explanation could due that in Escera, Yago, & Alho (2001) no significant distraction effects were observed between asynchronies. Actually, in the present study, short deviant stimuli which also no differed in RT, either did not reach significant difference in the RON peak-latency. It could mean that when distractor affects to the target processing in a similar way, the orienting of attention can be accomplished in a the same temporal window. However, whether the distractor affects different stages from the target processing, then the reorienting of attention to the task-relevant information will be also accomplished in different way from the impaired stage. In summary, the present study suggests, first, that classifying short stimuli, of 200 ms, and long stimuli, of 400 ms, take account different target processing. Second, the unexpected occurrence of a task- 81 irrelevant pitch change during this stimulus duration classification impaired performance. Third, the behavioral effects of task-irrelevant pitch changes depend on the impaired stage of the ongoing target processing. Fourth, the orienting of task-relevant information indexed by RON depended on the impaired stage of the target processing caused by the distractor. Acknowledgements The authors would like to thank Theda Ohlenbusch for her help in the selection of the sample. This study was supported by the Spanish Ministry of Education and Science (SEJ2006-00496/PSIC, HA20000022, and FP2000-5453), the Generalitat de Catalunya (SGR2005000954), the DAAD and the DFG-Working Memory Group of University of Leipzig. 82 Results References Berti, S., Roeber, U., & Schroger, E. (2004). Bottom-up influences on working memory: Behavioral and electrophysiological distraction varies with distractor strength. Experimental Psychology, 51(4), 249-257. Berti, S., & Schroger, E. (2003). Working memory controls involuntary attention switching: Evidence from an auditory distraction paradigm. The European Journal of Neuroscience, 17(5), 11191122. Escera, C., Alho, K., Winkler, I., & Naatanen, R. (1998). Neural mechanisms of involuntary attention to acoustic novelty and change. Journal of Cognitive Neuroscience, 10(5), 590-604. Escera, C., Corral, M. J., & Yago, E. (2002). An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, duration and intensity changes. Brain Research.Cognitive Brain Research, 14(3), 325-332. Escera, C., Yago, E., & Alho, K. (2001). Electrical responses reveal the temporal dynamics of brain events during involuntary attention switching. The European Journal of Neuroscience, 14(5), 877-883. Jankowiak, S., & Berti, S. (2007). Behavioral and event-related potential distraction effects with regularly occurring auditory deviants. Psychophysiology, 44(1), 79-85. McCarthy, G. & Wood, C.C. (1985). Scalp distributions of eventrelated potentials: an ambiguity associated with analysis of variance models. Electroencephalography and Clinical Neurophysiololgy, 62, 203-208. Munka, L., & Berti, S. (2006). Examining task-dependencies of different attentional processes as reflected in the P3a and reorienting negativity components of the human event-related brain potential. Neuroscience Letters, 396(3), 177-181. Escera, C., Yago, E., Corral, M. J., Corbera, S., & Nunez, M. I. (2003). Attention capture by auditory significant stimuli: Semantic analysis follows attention switching. The European Journal of Neuroscience, 18(8), 2408-2412. Naatanen, R. (1992). Attention and Brain Function. Erlbaum, Hillsdale, NJ. 83 Parmentier, F. B., Elford, G., Escera, C., Andres, P., & San Miguel, I. (2008). The cognitive locus of distraction by acoustic novelty in the cross-modal oddball task. Cognition, 106(1), 408-432. Rinne, T., Sarkka, A., Degerman, A., Schroger, E., & Alho, K. (2006). Two separate mechanisms underlie auditory change detection and involuntary control of attention. Brain Research, 1077(1), 135-143. Roeber, U., Berti, S., & Schroger, E. (2003). Auditory distraction with different presentation rates: An event-related potential and behavioral study. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 114(2), 341349. Roeber, U., Widmann, A., & Schroger, E. (2003). Auditory distraction by duration and location deviants: A behavioral and event-related potential study. Brain Research.Cognitive Brain Research, 17(2), 347-357. Schroger, E. (1996). Neural mechanism for involuntary attention shifts to changes in auditory stimulation. Journal of Cognitive Neuroscience, 8, 527-539. Schroger, E., Giard, M. H., & Wolff, C. (2000). Auditory distraction: Event-related potential and behavioral indices. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 111(8), 1450-1460. Schroger, E., & Wolff, C. (1998a). Behavioral and electrophysiological effects of task-irrelevant sound change: A new distraction paradigm. Brain Research.Cognitive Brain Research, 7(1), 71-87. Schroger, E., & Wolff, C. (1998b). Attentional orienting and reorienting is indicated by human event-related brain potentials. Neuroreport, 9(15), 3355-3358. Sussman, E., Winkler, I., & Schroger, E. (2003). Top-down control over involuntary attention switching in the auditory modality. Psychonomic Bulletin & Review, 10(3), 630-637. 84 Results Figure 1. Illustration of the stimuli. All stimuli were targets, with most of them (88%) with the same pitch during their whole length (standards tones, STD), and some of them (12%) including a pitch change (deviant stimuli) at 50 ms (early-DEV) or at 150 ms (late-DEV) from stimulus onset. Arrow indicated the target onset. 85 Figure 2. Mean error rate (ER) and response time (RT) for standard tones (STD) and deviant stimuli (early-DEV or late-DEV), for short (black) and long (grey) stimuli. Bars indicate the standard error of the mean. Because hit rate and error rate analysis showed analogous results, just ER has been plotted. 86 Results Figure 3. Left, grandaverage ERPs elicited to frequent standard (STD) and to infrequent deviant, early-DEV (e-DEV) and late-DEV (l-DEV) for short (dot line) and long stimuli (plain line). Note that the orienting of attention indexed by P3a peaked at different temporal window of the target stimulus processing. Specifically, when the distractor feature occurred far away from the target onset (e-DEV), the P3a overlapped the N2 standard window, whereas when the distractor feature occurred closer to target onset (l-DEV), the P3a peaked over the raising slope of the standard N2 ERP and shortly preceding the target feature P3b window. Right, scalp isopotential maps around 10 ms of the P3b peak for each target. 87 Figure 4. Left, grand-average difference waveforms obtained by subtracting the ERPs elicited by the standard stimuli from those elicited to the deviant stimuli at Fz. Right, scalp isopotential maps of the P3a and RON component around the analyzed peak-latency window. 88 General discussion 5. General discussion Participants classified odd/even or short/long stimuli with an overall hit rate of above 85%. However, random changes embedded in the repetitive stream of auditory stimulation impaired performance. Responses were from 10 ms to 72 ms slower when they were preceded by a sound change than when they were preceded by a repetitive sound, a phenomenon referred to as distraction. Different types of changes to the physical features of the sound were able to cause distraction. Slight modifications of intensity (7% softer), duration (75% shorter) or pitch (17% higher) regarding the standard stimulation produced similar hit rate decreases (5.2%), error rate increases (5.6%) and slower response times (24 ms) with regard to the first study. Moreover, slight pitch changes (10% higher) in the third study and complex (with a broad spectral range) novel environmental sounds in the second study produced slower responses (72 ms and 10 ms, respectively). Therefore, the brain system seems to detect slight or large task-irrelevant changes that break the preceding regularity. The evaluation of these unexpected irruptions presumably reallocated the attentional resources from the task-relevant to the task-irrelevant information, and impaired performance, as reflected by the behavioral results. The distraction pattern obtained for slight changes (deviant stimuli: first and third study) and novel sounds (second study) also differed. Hit and error rates were affected in the case of deviant stimuli, but not for novel sounds, as has been reported before (Escera et al., 1998, 2001). If we assume that 89 the classification of the stimulus is related to hit and error rate measures, it seems that novel sounds did not interfere with the first stages of the target processing. Actually, using novel sounds as task-irrelevant stimuli, Parmentier et al. (2008) showed that increasing the difficulty of the stimulus identification or the categorical processing of the target had no impact on the magnitude of the distraction. Only when a recapture signal was presented before the target was the distraction abolished. The authors interpreted that the slower responses produced by novel sounds (in the absence of recapture signal) were not due to any interference in the target processing, but to the moving of attention from the task-irrelevant to taskrelevant information. The next open question was to determine what kind of shifts could have been involved in these effects. Task-relevant information was visual and was presented circa 1 m from the participant’s eyes. Task-irrelevant information was auditory and presented via headphones placed on the participant’s head. So, at least two kinds of shifts could be involved: one between sensory modalities (from auditory to visual), and another between spatial locations (from headphones to the screen). The second study has helped to elucidate this question. In this study, the spatial distance between the task-relevant and the task-irrelevant channels was manipulated, while the two types of sensory modality remained constant. The results revealed that when task-relevant and task-irrelevant information shared the same location, the distraction was extinguished. However, in the rest of the conditions, where task-relevant and taskirrelevant were presented at different locations, distraction did not follow a pattern that was exactly proportional to distance. Novel sounds presented from the right hemispace of the participant caused distraction, while novel sounds occurring at homologous locations on the left hemispace did not impair performance. These results were in agreement with a series of studies 90 General discussion on involuntary orienting of attention in the visual modality, which showed larger disturbing effects to pop-out or prime stimuli occurring in the right visual than in the left visual hemifield (Castro-Barros et al., 2008; Pollmann, 1996, 2000). In addition, the magnetic counterpart of the MMN elicited by location deviant stimuli showed that the right but not the left lateralized deviants elicited equally short mismatch latencies in the two cerebral hemispheres, suggesting an advantage for processing auditory spatial information in the right hemispace than in the left one (Kaiser et al., 2000). Taking all these findings together, a rightward bias for the behavioral influence of unexpected stimuli in the right side of the environment it seems plausible (Mesulam, 1990). Therefore, the spatial location of the auditory task-irrelevant and visual task-relevant stimuli becomes fundamental to the attention system, supporting the notion that the distraction effects of novel sounds could be caused by a spatial shifting of attention. Besides affecting response time, slight deviant sounds also produced hit rate decreases and error rate increases (first and third study). Actually, the third study showed that responses to the long stimuli produced more errors when the deviance onset was far away from the target onset (150 ms) than when it was nearby (50 ms). Again, if we assume that stimulus classification could be involved in the hit rate and error rate measures, it appears that slight deviant changes could be able to affect the first stages of the target processing. Moreover, the opposite pattern of results to the response time for the long stimuli was recorded. When the temporal distance between deviance and target onset was longer (150 ms), faster responses were recorded, while for the shorter distance responses were slower. So, in this case the response time was more affected when the hit rate and error rate were more preserved. These apparently contradictory results make sense if we assume that task-relevant (duration) and task-irrelevant (pitch) features are processed in parallel at the first stages. Therefore, when the distance between the deviance and the target onset was shorter, the classification 91 processing was presumably more advanced, yielding better hit and error rates, but impairing the later stage of the target processing associated with the response processing (causing slower response times). However, when the distance between the deviance and target onset was longer, the classification processing could not be finished; this would have influenced hit or error rate measures, but would have had a lesser effect on response time. Another result that supports this view is the difference in distracting effects found for the two types of targets used in the auditory-auditory distraction paradigm. While the long stimuli yielded the behavioral results discussed above, short stimuli did not show any statistically significant differences in hit or error rates for any deviance onset. However, the response time to short stimuli was faster in general than to long stimuli, which may indicate a different underlying cognitive processing for resolving the two types of target. So, if classifying short and long stimuli requires different target processing, it seems plausible that a deviant feature occurring in two different temporal windows could also affect different stages of the target processing for short and long stimuli, which would be reflected by different distracting effects, precisely the outcome shown here. Because responses to short stimuli were faster, deviance onset further away or closer to the target onset was not able to affect the classification stimulus processing. However, short deviant stimuli caused similarly slower responses for both deviance onsets, suggesting that when the target processing was faster, the deviant feature affected a similar stage of the target processing. In summary, while the distracting effects observed by novel sounds were due to a total allocation of attentional resources before the target processing, slight deviant changes seem to allow a parallel processing of the taskrelevant and task-irrelevant features. One could argue that the asynchrony used in the second study between the novel sounds and the visual stimuli was too long (300 ms) to impair the classification stage of the processing. 92 General discussion However, deviant stimuli were presented with the same asynchrony in the first study and, even so, hit rate and error rate appeared affected. Therefore, although we cannot conclusively rule out the possibility that novel sounds presented in short asynchronies as those of the third study could impair classification measures, at present it seems that novel sounds recapture attentional resources in a different way from deviant sounds. The electrophysiological activity recorded during the presentation of these stimuli also helped to elucidate the distraction effects. The difference waves obtained from the subtraction of the standard ERPs from the deviant ERPs allowed us to isolate the neural activity due to distraction. All the types of slight deviant changes (first and third study) elicited a first negative component around 150-200 ms identified as MMN, which corroborates its role as an early automatic call for focal attention. In the case of novel sounds with a rich frequency range, its elicitation was subordinate the N1 component, which is usually triggered by abrupt changes in the background stimulation. Obviously, the scalp distribution of MMN varied for each type of change (pitch, duration or intensity) suggesting that specific neural populations were involved in the detection of different auditory features, in agreement with previous studies (Giard et al., 1994). Following the MMN, a positive component called P3a was elicited for novel (second study) and slight pitch changes (third study). Two results from the present thesis have broadened our understanding of P3a. First, its amplitude is sensitive to the spatial location between task-relevant and task-relevant information (second study). That is, P3a amplitude was larger at closer locations between novel sound distractors and visual targets. Second, its scalp distribution and its temporal peak were affected by the ongoing stage of target processing (third study). P3a was elicited faster when the taskirrelevant and task-relevant were closer (50 ms) than when it was far away (150 ms). Also, P3a presented a large central positivity distribution when it 93 was elicited during the first stages of the target processing. In summary, P3a generation was sensitive to the spatial location of the unexpected changes and the underlying ongoing cognitive processing; this latter finding is also in agreement with previous and related studies (Dominguez-Borras et al., 2008a; San Miguel et al., 2008a). Finally, the reorienting negativity (RON) following the P3a ERP component on distraction context was not absolutely independent on the deviance onset (third study). In a previous study, Escera et al. (2001) observed that RON peaked to a similar time window to different distractor-to-target asynchronies (245 and 355 ms). Their results confirmed the role of RON as index of the attentional allocation to the processing of the relevant information after a momentary distraction (Schroger and Wolf, 1998b). Nevertheless, the shorter distractor-to-target asynchronies (150 and 50 ms) used in the third study of the present thesis did not confirm this result on the auditory-auditory paradigm. A possible explanation could be that the intervals used by Escera et al. (2001) allowed the orientation of attention on a similar time window. Actually, no behavioral differences between the two types of asynchronies were observed. However, when the asynchrony between the distractor and the target was shorter (third study), significant behavioral differences were obtained. Therefore, returning to primary task performance after impairing the classification or response stage of target processing should exhibit a different cognitive process, as shown by the temporal and scalp distribution of RON results. While the present thesis has resolved some issues on the brain’s control of attention, new questions have emerged. Further investigation should explore the attentional mechanisms activated by novel sounds using shorter distractor-to-target intervals. If novel sounds do not really allow parallel processing, hit and error rate measures would still be unaffected. Also, an independent analysis of the types of target in the auditory-visual paradigm 94 General discussion could reveal deviant and target processing interaction. If classifying one type of target differs from the others, specific patterns of behavioral and electrophysiological measures of distraction should also be found. On the other hand, many studies in the spatial cross-modal attentional literature (Eimer & Schroger, 1998; Eimer, 1999; Spence & Read, 2003) have shown a P3a-like positivity when the target stimulus appeared in an unexpected location. Closer investigations involving distraction and spatial attention studies could help to elucidate the cognitive role of P3a in attention. And, because the distraction potential (DP) has been recorded in the absence of behavioral distraction (condition 0º in the third study, Munka & Berti, 2006; Polo et al. 2003), new studies should continue to review the underlying cognitive process of its components as has recently been suggested (Barcelo et al., 2006; Berti, 2008a). 95 Conclusions 6. Conclusions The present thesis investigated the electrophysiological indexes of the detection and processing of auditory distractors. The resulting conclusions can be summarized as follows: ƒ Across all the studies, slightly deviant and novel sounds embedded in a repetitive stream of auditory stimulation caused distraction (slower response times). The distraction pattern of slightly deviant and novel sounds differed from each other. Slightly deviant and target features could be processed in parallel, which affected the classification processing (i.e., reductions in hit rate and error rate) and delay response time. In turn, novel sounds seem to produce a total reallocation of attentional sources before the beginning of the target processing, delaying the response time, but with no impairment in any stage of the target processing. ƒ The results of the second study showed that spatial location of the sounds was a relevant factor for distraction. When task-irrelevant stimuli were situated in the right hemifield far away from the taskrelevant stimulus, distraction was observed. However, when the taskirrelevant and task-relevant stimulus shared the same location or when the task-irrelevant stimulus was situated in the left hemifield, distraction was abolished. ƒ When the target processing was long-lasting (long stimuli), the manipulation of the temporal distance between the distractor and the target onset in the third study also became a relevant parameter for the pattern of distraction. Impaired impulsive response style was observed in the case of the longest interval (150 ms) between the 97 distractor and the target feature, while impaired reflexive response style was obtained in the case of the shortest interval (50 ms). ƒ The elicitation of mismatch negativity (MMN) for pitch, duration and intensity changes in the first study confirmed its role as a genuine change detector mechanism. ƒ The attentional switch from the task-relevant to the task-irrelevant information identified in the ERP components as P3a was sensitive to the distractor location in the second study. Larger mean amplitude was found in the case of closer locations between the task-irrelevant and the task-relevant information. So the attentional switch indexed by P3a could indeed be a switch of the attentional resources between spatial locations. ƒ The cognitive processing of the return to primary task performance after a momentary distraction indicated by the reorienting negativity (RON) depends, according to the third study, on the impaired stage of the distractor to the ongoing target processing. 98 References 7. References Alho, K., Escera, C., Diaz, R., Yago, E., & Serra, J. M. (1997). Effects of involuntary auditory attention on visual task performance and brain activity. Neuroreport, 8(15), 3233-3237. Andres, P., Parmentier, F. B., & Escera, C. (2006). The effect of age on involuntary capture of attention by irrelevant sounds: A test of the frontal hypothesis of aging. Neuropsychologia, 44(12), 2564-2568. Barcelo, F., Escera, C., Corral, M. J., & Perianez, J. A. (2006). Task switching and novelty processing activate a common neural network for cognitive control. Journal of Cognitive Neuroscience, 18(10), 1734-1748. Berti, S. (2008a). Cognitive control after distraction: Event-related brain potentials (ERPs) dissociate between different processes of attentional allocation. Psychophysiology, in press. Berti, S. (2008b). Object switching within working memory is reflected in the human event-related brain potential. Neuroscience Letters, 434(2), 200205. Berti, S., Roeber, U., & Schroger, E. (2004). Bottom-up influences on working memory: Behavioral and electrophysiological distraction varies with distractor strength. Experimental Psychology, 51(4), 249-257. Berti, S., & Schroger, E. (2003). Working memory controls involuntary attention switching: Evidence from an auditory distraction paradigm. The European Journal of Neuroscience, 17(5), 1119-1122. Castro-Barros, B. A., Righi, L. L., Grechi, G., & Ribeiro-do-Valle, L. E. (2008). Interlateral asymmetry in the time course of the effect of a peripheral prime stimulus. Brain and Cognition, 66(3), 265-279. Corbera, S. & Escera, C. (2008). Increased distractibility in developmental dyslexia as revealed by behavioral and event-related brain potential measures, in preparation. Cortinas, M., Corral, M. J., Garrido, G., Garolera, M., Pajares, M., & Escera, C. (2008). Reduced novelty-P3 associated with increased behavioral distractibility in schizophrenia. Biological Psychology, 78(3), 253-260. Deouell, L.Y. (2007). The frontal generator of the mismatch negativity revisited. Journal of Psychophysiology, 21(3-4), 188-203. 99 Dominguez-Borras, J., Garcia-Garcia, M., & Escera, C. (2008a). Negative emotional context enhances auditory novelty processing. Neuroreport, 19(4), 503-507. Dominguez-Borras, J., Trautmann, S., Fehr, T., Ehrard, P. & Herrmann, M. & Escera, C. (2008b). Gating of auditory novelty processing by emotional context in superior temporal gyrus, submitted. Dominguez-Borras, J., Garcia-Garcia, M., & Escera, C. (2008c). Emotional context enhances auditory novelty processing: behavioural and electrophysiological evidence, submitted. Eimer, M. (1999). Can attention be directed to opposite locations in different modalities? an ERP study. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 110(7), 1252-1259. Eimer, M., & Driver, J. (2001). Crossmodal links in endogenous and exogenous spatial attention: Evidence from event-related brain potential studies. Neuroscience and Biobehavioral Reviews, 25(6), 497-511. Eimer, M., & Schroger, E. (1998). ERP effects of intermodal attention and cross-modal links in spatial attention. Psychophysiology, 35(3), 313-327. Escera, C. (2007). The mismatch negativity 30 years later: How far have we come? Journal of Psychophysiology, 21(3-4), 129-132. Escera, C., Alho, K., Schroger, E., & Winkler, I. (2000). Involuntary attention and distractibility as evaluated with event-related brain potentials. Audiology & Neuro-Otology, 5(3-4), 151-166. Escera, C., Alho, K., Winkler, I., & Naatanen, R. (1998). Neural mechanisms of involuntary attention to acoustic novelty and change. Journal of Cognitive Neuroscience, 10(5), 590-604. Escera, C., & Corral, M. J. (2003). The distraction potential (DP), an electrophysiological tracer of involuntary attention control and its dysfunction. In I. Reinvang, M. W. Greenlee & M. Herrmann (Eds.), The cognitive neuroscience of individual differences (pp. 63-76). Oldenburg: Bibliotheks-und Informationssystem der Universität Oldenburg. Escera, C., & Corral, M. J. (2007). Role of mismatch negativity and novelty-P3 in involuntary auditory attention. Journal of Psychophysiology, 21(3-4), 251-264. Escera, C., Corral, M. J., & Yago, E. (2002). An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, 100 References duration and intensity changes. Brain Research.Cognitive Brain Research, 14(3), 325-332. Escera, C., Yago, E., & Alho, K. (2001). Electrical responses reveal the temporal dynamics of brain events during involuntary attention switching. The European Journal of Neuroscience, 14(5), 877-883. Escera, C., Yago, E., Corral, M. J., Corbera, S., & Nunez, M. I. (2003). Attention capture by auditory significant stimuli: Semantic analysis follows attention switching. The European Journal of Neuroscience, 18(8), 24082412. Friedman, D., Cycowicz, Y. M., & Gaeta, H. (2001). The novelty P3: An eventrelated brain potential (ERP) sign of the brain's evaluation of novelty. Neuroscience and Biobehavioral Reviews, 25(4), 355-373. Giard, M. H., Perrin, F., Echallier, J. F., Thevenet, M., Froment, J. C., & Pernier, J. (1994). Dissociation of temporal and frontal components in the human auditory N1 wave: A scalp current density and dipole model analysis. Electroencephalography and Clinical Neurophysiology, 92(3), 238-252. Gumenyuk, V., Korzyukov, O., Alho, K., Escera, C., & Naatanen, R. (2004). Effects of auditory distraction on electrophysiological brain activity and performance in children aged 8-13 years. Psychophysiology, 41(1), 30-36. Gumenyuk, V., Korzyukov, O., Alho, K., Escera, C., Schroger, E., Ilmoniemi, R. J., et al. (2001). Brain activity index of distractibility in normal school-age children. Neuroscience Letters, 314(3), 147-150. Gumenyuk, V., Korzyukov, O., Escera, C., Hamalainen, M., Huotilainen, M., Hayrinen, T., et al. (2005). Electrophysiological evidence of enhanced distractibility in ADHD children. Neuroscience Letters, 374(3), 212-217. Heinze, H. J., Mangun, G. R., Burchert, W., Hinrichs, H., Scholz, M., Munte, T. F., et al. (1994). Combined spatial and temporal imaging of brain activity during visual selective attention in humans. Nature, 372(6506), 543-546. Herrmann, C.S. & Knight, R.T. (2001). Mechanisms of human attention: eventrelated potentials and oscillations. Neuroscience & Biobehavioral Reviews, 25, 465-476. Horvath, J., Maess, B., Berti, S., & Schroger, E. (2008). Primary motor area contribution to attentional reorienting after distraction. Neuroreport, 19(4), 443-446. 101 Jaaskelainen, I., Alho, K., Escera, C., Winkler, I., Sillanaukee, P., & Naatanen, R. (1996). Effects of ethanol and auditory distraction on forced choice reaction time. Alcohol, 13, 153-156. Jaaskelainen, I. P., Schroger, E., & Naatanen, R. (1999). Electrophysiological indices of acute effects of ethanol on involuntary attention shifting. Psychopharmacology, 141(1), 16-21. Jacobsen, T., & Schroger, E. (2001). Is there pre-attentive memory-based comparison of pitch? Psychophysiology, 38(4), 723-727. Jankowiak, S., & Berti, S. (2007). Behavioral and event-related potential distraction effects with regularly occurring auditory deviants. Psychophysiology, 44(1), 79-85. Kaiser, J., Lutzenberger, W., & Birbaumer, N. (2000). Simultaneous bilateral mismatch response to right- but not leftward sound lateralization. Neuroreport, 11(13), 2889-2892. Knight, R. T. (1984). Decreased response to novel stimuli after prefrontal lesions in man. Electroencephalography and Clinical Neurophysiology, 59(1), 9-20. McCarthy, G. & Wood, C.C. (1985). Scalp distributions of event-related potentials: an ambiguity associated with analysis of variance models. Electroencephalography and Clinical Neurophysiololgy, 62, 203-208. Mesulam, M. M. (1990). Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Annals of Neurology, 28(5), 597-613. Munka, L., & Berti, S. (2006). Examining task-dependencies of different attentional processes as reflected in the P3a and reorienting negativity components of the human event-related brain potential. Neuroscience Letters, 396(3), 177-181. Naatanen, R. (1990). The role of attention in auditory information processing as revealed by event-related potentials and other brain measures of cognitive function. Behavioral and Brain Sciences, 13, 201-288. Naatanen, R., (2007). The mismatch negativity (MMN): Where is the big fish? Journal of Psychophysiology, 21(3-4), 133-137. Naatanen, R., Paavilainen, P., Alho, K., Reinikainen, K., & Sams, M. (1989). Do event-related potentials reveal the mechanism of the auditory sensory memory in the human brain? Neuroscience Letters, 98(2), 217-221. 102 References Naatanen, R., Paavilainen, P., & Reinikainen, K. (1989). Do event-related potentials to infrequent decrements in duration of auditory stimuli demonstrate a memory trace in man? Neuroscience Letters, 107(1-3), 347352. Naatanen, R., & Picton, T. (1987). The N1 wave of the human electric and magnetic response to sound: A review and an analysis of the component structure. Psychophysiology, 24(4), 375-425. Parmentier, F. B., Elford, G., Escera, C., Andres, P., & San Miguel, I. (2008). The cognitive locus of distraction by acoustic novelty in the cross-modal oddball task. Cognition, 106(1), 408-432. Pollmann, S. (1996). A pop-out induced extinction-like phenomenon in neurologically intact subjects. Neuropsychologia, 34(5), 413-425. Pollmann, S. (2000). Extinction-like effects in normals: Independence of localization and response selection. Brain and Cognition, 44(3), 324-341. Polo, M. D., Escera, C., Yago, E., Alho, K., Gual, A., & Grau, C. (2003). Electrophysiological evidence of abnormal activation of the cerebral network of involuntary attention in alcoholism. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 114(1), 134-146. Polo, M. D., Newton, P., Rogers, D., Escera, C., & Butler, S. (2002). ERPs and behavioural indices of long-term preattentive and attentive deficits after closed head injury. Neuropsychologia, 40(13), 2350-2359. Rinne, T., Sarkka, A., Degerman, A., Schroger, E., & Alho, K. (2006). Two separate mechanisms underlie auditory change detection and involuntary control of attention. Brain Research, 1077(1), 135-143. Roeber, U., Berti, S., & Schroger, E. (2003). Auditory distraction with different presentation rates: An event-related potential and behavioral study. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 114(2), 341-349. Roeber, U., Widmann, A., & Schroger, E. (2003). Auditory distraction by duration and location deviants: A behavioral and event-related potential study. Brain Research.Cognitive Brain Research, 17(2), 347-357. San Miguel, I., Corral, M. J., & Escera, C. (2008a). When loading working memory reduces distraction: Behavioral and electrophysiological evidence from an auditory-visual distraction paradigm. Journal of Cognitive Neuroscience, in press. 103 San Miguel, I., Escera, C., Erhard, P., Fehr, T., & Herrmann, M. (2008b). Working memory load and distraction interaction: an fMRI investigation, in preparation. Schroger, E. (1996). Neural mechanism for involuntary attention shifts to changes in auditory stimulation. Journal of Cognitive Neuroscience, 8, 527-539. Schroger, E. (2007). Mismatch negativity: a microphone into auditory memory. Journal of Psychophysiology, 21(3-4), 138-146. Schroger, E., Giard, M. H., & Wolff, C. (2000). Auditory distraction: Eventrelated potential and behavioral indices. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 111(8), 1450-1460. Schroger, E., & Wolff, C. (1998a). Behavioral and electrophysiological effects of task-irrelevant sound change: A new distraction paradigm. Brain Research.Cognitive Brain Research, 7(1), 71-87. Schroger, E., & Wolff, C. (1998b). Attentional orienting and reorienting is indicated by human event-related brain potentials. Neuroreport, 9(15), 3355-3358. Simons, R. F., Graham, F. K., Miles, M. A., & Chen, X. (2001). On the relationship of P3a and the novelty-P3. Biological Psychology, 56(3), 207218. Spence, C. (2002). Multisensory attention and tactile information-processing. Behavioural Brain Research, 135(1-2), 57-64. Spence, C., & Read, L. (2003). Speech shadowing while driving: On the difficulty of splitting attention between eye and ear. Psychological Science: A Journal of the American Psychological Society / APS, 14(3), 251-256. Squires, N. K., Squires, K. C., & Hillyard, S. A. (1975). Two varieties of longlatency positive waves evoked by unpredictable auditory stimuli in man. Electroencephalography and Clinical Neurophysiology, 38(4), 387-401. Sussman, E., Winkler, I., & Schroger, E. (2003). Top-down control over involuntary attention switching in the auditory modality. Psychonomic Bulletin & Review, 10(3), 630-637. Winkler, I. (2007). Interpreting the mismatch negativity (MMN). Journal of Psychophysiology, 21(3-4), 147-163. 104 References Wetzel, N., Berti, S., Widmann, A., & Schroger, E. (2004). Distraction and reorientation in children: A behavioral and ERP study. Neuroreport, 15(8), 1355-1358. Wetzel, N., & Schroger, E. (2007a). Cognitive control of involuntary attention and distraction in children and adolescents. Brain Research, 1155, 134146. Wetzel, N., & Schroger, E. (2007b). Modulation of involuntary attention by the duration of novel and pitch deviant sounds in children and adolescents. Biological Psychology, 75(1), 24-31. Wetzel, N., Widmann, A., Berti, S., & Schroger, E. (2006). The development of involuntary and voluntary attention from childhood to adulthood: A combined behavioral and event-related potential study. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 117(10), 2191-2203. Yago, E., Corral, M. J., & Escera, C. (2001). Activation of brain mechanisms of attention switching as a function of auditory frequency change. Neuroreport, 12(18), 4093-4097. Yago, E., Escera, C., Alho, K., & Giard, M. H. (2001). Cerebral mechanisms underlying orienting of attention towards auditory frequency changes. Neuroreport, 12(11), 2583-2587. Yago, E., Escera, C., Alho, K., Giard, M. H., & Serra-Grabulosa, J. M. (2003). Spatiotemporal dynamics of the auditory novelty-P3 event-related brain potential. Brain Research.Cognitive Brain Research, 16(3), 383-390. 105 Appendixes Catalan summary of the thesis Resum de la tesi Títol Indicadors electrofisiològics de la detecció i el processament de distractors auditius Introducció A la vida diària, els nostres sentits estan contínuament bombardejats per informació procedent de diferents modalitats i localitzacions. Per aconseguir concentrar-nos en les nostres tasques necessitem la capacitat de seleccionar els estímuls rellevants d’entre tota la informació que ens envolta. Tot i així, un bloqueig total de qualsevol estímul irrellevant resultaria inadaptatiu per la incapacitat de poder reaccionar a canvis potencialment importants com, per exemple, la percepció del plor d’un nadó quan ens trobem en espais diferents de la casa, o la ruptura d’un vidre prop nostre. Estudis recents han investigat com el sistema nerviós porta a terme la selecció de la informació rellevant per la tasca en curs, alhora que permet l’avaluació de possibles canvis auditius en l’entorn. 109 Paradigmes de distracció auditiva Diversos paradigmes han intentat recrear al laboratori la interferència provocada per canvis en l’entorn acústic. Una de les possibles tasques encomanda als participants consisteix en classificar estímuls auditius, curts o llargs (Schroger & Wolff, 1998a), o bé, estímuls visuals, parells o senars (Escera, Alho, Winkler, & Naatanen, 1998). Aquests estímuls rellevants són precedits per una estimulació auditiva constant. De forma infreqüent i inesperada l’estimulació constant és reemplaçada per un altre tipus de so. Anomenem estímuls auditius discrepants quan es modifica lleugerament alguna característica física respecte l’estimulació auditiva precedent, com pot ser la freqüència, la intensitat o la durada. En canvi, quan l’estimulació constant és reemplaçada per estímuls únics o totalment diferents, com pot ser el so d’un telèfon o el caure d’unes claus, parlem d’estímuls innovadors (Escera et al., 1998). Ambdós tipus de canvis provoquen respostes lleugerament més lentes (de l’ordre de milisegons), fenomen identificat com a distracció. Factors que influeixen en la distracció auditiva Dos dels paràmetres coneguts que afecten la distracció són la magnitud del canvi (Berti, Roeber, & Schroger, 2004) i la seva predrictibilitat (Sussman, Winkler, & Schroger, 2003). És a dir, grans i impredictibles canvis, tal com el so d’un telèfon, òbviament produeixen major distracció, que canvis molt similars a l’estimulació precedent o que poden ser previstos per altres estímuls (Escera et al., 1998; Escera, Yago, & Alho, 2001). La magnitud de la distracció s’ha observat proporcional al canvi de to entre l’estímul repetitiu i l’estímul discrepant (Berti et al., 2004; Jaaskelainen, Schroger, & Naatanen, 1999; Schroger, 1996). Seguint la mateixa línia, estímuls innovadors susceptibles de ser reconeguts, com per exemple, el timbre del telèfon 110 o el soroll d’una perforadora al carrer, generen respostes més lentes que estímuls innovadors (sons complexes amb un ample espectre de freqüència) difícilment associats a cap element conegut (Escera, Yago, Corral, Corbera, & Nunez, 2003). La predictibilitat dels estímuls discrepants ha estat estudiat per Sussman et al. (2003). En aquesta investigació, cada so estava precedit per un estímul visual. Dues condicions van ser presentades. En la condició previsible, l’estímul visual indicava la freqüència del so. En la condició imprevisible, l’estímul visual es presentava aleatòriament emparellat amb la freqüència del so, sense proporcionar cap informació sobre la freqüència de l’estímul subseqüent. Els estímuls discrepants de la condició impredictible van provocar distracció, però els estímuls discrepants de les series predictibles no. Causa en la distracció auditiva Per quin motiu els canvis en l’entorn acústic provoquen un pitjor rendiment de la tasca en curs? Una possible explicació rau en que part dels recursos atencionals destinats a l’execució de la tasca encomada s’utilitzin per l’avaluació del nou estímul. Però, en quin moment el canvi irromp en el processament dels estímuls rellevants? Alho, Escera, Diaz, Yago, & Serra, (1997) van observar que el potencial evocat N1 de l’estímul visual rellevant es veia atenuat quan era precedit per un estímul discrepant, en comparació a quan era precedit per un estímul repetitiu. Aquest resultat va suggerir als autors que el distractor era capaç d’interrompre una fase primerenca del processament de l’estímul rellevant. Més tard, Parmentier, Elford, Escera, Andres, & San Miguel, (2008) van voler corroborar aquest resultat en el cas d’estímuls innovadors. La magnitud de la distracció no es va veure afectada ni per la dificultat de discriminació visual de l’estímul rellevant, ni per l’augment de categories a l’hora de classificar els estímuls rellevants per la tasca. Només en una condició en la qual es presentava un 111 estímul captador de l’atenció, just abans de l’aparició de l’estímul rellevant, va ser capaç d’anular la distracció conductual. Per tant, la distracció en el cas d’estímuls auditius no semblava estar causada per una interferència en etapes primerenques o tardanes del processament de l’estímul rellevant, sinó pel moviment de l’atenció del distractor (estímul innovador) fins l’estímul rellevant. La següent qüestió oberta a preguntar-se és de quina naturalesa resulta aquest moviment. Els resultats de la present tesi (estudi segon) permetran resoldre si es tracta d’un moviment en l’espai (de la localització en la què es presenten els distractors –a través dels auriculars– fins a la pantalla per on es presenten els estímuls visuals rellevants –situada aproximadament a un metre del participant), o bé, si el moviment es tracta de traspassar els recursos atencionals destinats a la modalitat visual (rellevant) cap a la modalitat auditiva (irrellevant). Finestral temporal vulnerable a la distracció auditiva Schröger (1996) va observar un empitjorament de les respostes conductuals (decrement del nombre de respostes correctes i augment del temps de resposta) quan el distractor es presentava 200 ms abans de l’estímul rellevant, però no, quan es presentava a 560 ms. Posteriorment, Escera et al. (2001) va utilitzar també dues asincronies entre l’estímul distractor i l’estímul rellevant: a 245 ms i a 355 ms. Ambdós intervals van provocar respostes més lentes, encara que no van mostrar diferència significativa ni quan el distractor es tractava d’un estímul discrepant, ni quan es tractava d’un estímul innovador. Altres estudis han registrat distracció conductual utilitzat intervals de 200 ms (Roeber, Berti, & Schroger, 2003; Roeber, Widmann, & Schroger, 2003; Schroger & Wolff, 1998b; Schroger, Giard, & Wolff, 2000), 100 ms (Schroger & Wolff, 1998a) o, fins i tot, quan l’estímul distractor i l’estímul discrepant es presentaven a l’hora (Rinne, Sarkka, Degerman, Schroger, & Alho, 2006). Un dels altres objectius de la 112 present tesi (tercer estudi) és esbrinar si la distància temporal entre el distractor i l’estímul rellevant resulta decisòria per la magnitud de la distracció. Electrofisiologia de la distracció auditiva Els potencials evocats registrats durant l’execució de les tasques de distracció permeten investigar la dinàmica espacial i temporal de les xarxes neuronals implicades en el control de l’atenció. El potencial evocat típic en aquesta situacions mostra una complexa morfologia formada per l’activitat associada a l’estímul distractor i l’estímul rellevant. La resta aritmètica entre el potencial evocat davant de l’estímul distractor i el potencial evocat davant l’estímul repetitiu permet aïllar l’activitat cerebral relacionada amb la distracció conductual. Mitjançant aquesta simple operació, el potencial evocat resultant mostra típicament tres components que han estat relacionats amb diferents fases neurofisilògiques de la distracció. Primer, un component negatiu que apareix al voltant dels 150-200 ms des de l’aparició del distractor i que ha estat relacionat amb la detecció automàtica del canvi. Aquest component primerenc rep el nom de potencial de disparitat (mismatch negativity, MMN, en terminologia anglosaxona). Segon, una deflexió positiva que segueix al potencial de disparitat i que ha estat relacionada amb l’orientació efectiva de l’atenció cap al canvi, batejada amb el nom de P3a. Finalment, un component negatiu, identificat en terminologia anglosaxona com a reorienting negativity (RON), implicat amb la reorientació de l’atenció cap a l’estímul rellevant després d’una distracció momentània. Tot i així, alguns treballs mostren l’aparició d’aquests components en absència de distracció conductual (Munka & Berti, 2006; Polo et al., 2003), pel que una revisió independent per cada un d’ells pot ajudar a descriure més acuradament els processos cognitius subjacents a la seva generació. 113 La detecció del canvi i el potencial de disparitat (MMN) La detecció automàtica d’estímuls potencialment rellevants que ocorren fora del focus d’atenció ha estat relacionada amb, almenys, dos mecanismes cerebrals. Un dels mecanismes és activat per inicis o acabaments inesperats d’estímuls, com per exemple, la llum d’un llamp, l’alarma d’una sirena o el cessament del funcionament del motor de la nevera després de tot un dia en marxa. Un altre mecanisme diferent és activat quan un estímul en particular trenca amb la petja neural de la constant estimulació auditiva precedent. Un exemple típic de l’activació d’aquest mecanisme l’han utilitzat durant anys les emissores de ràdio per anunciar les notícies: una sèrie de tons constants finalitza amb un to més agut o més llarg. El primer mecanisme està basat en la reacció neurofisiològica als momentanis increments o decrements d’energia física i ha estat associat amb el component auditiu N1 (Naatanen & Picton, 1987). El segon mecanisme implica una avaluació constant de l’entorn acústic capaç de detectar qualsevol canvi que trenqui amb la regularitat precedent, i ha estat relacionat amb el potencial de disparitat (Naatanen, 1990, 2007; Schroger, 2007; Winkler, 2007). Un ampli nombre de treballs han indicat que tant el component N1, com el potencial de disparitat, tenen les seves fonts generadores al planum temporale de l’escorça auditiva (Alho, 1995; Alho et al., 1998; Escera, Alho, Schroger, & Winkler, 2000; Naatanen & Picton, 1987), amb contribucions de regions prefrontals (veure Giard et al., 1994 per N1; Deouell, 2007, pel potencial de disparitat). Tot i així, la majoria d’estudis acostumen a utilitzar com a distractors un canvi en el to de l’estimulació. Per tant, no és possible descartar totalment que la resposta cerebral associada a la detecció del canvi no estigui provocada per la resposta específica de neurones sensibles a 114 una determinada freqüència (Jacobsen & Schroger, 2001; Yago, Escera, Alho, & Giard, 2001). Estudis subseqüents, entre ells el primer estudi de la present tesi, han demostrat que, efectivament, el detector de canvis associat al potencial de disparitat també és sensible a modificacions en la durada (Escera, Corral, & Yago, 2002; Roeber et al., 2003), la intensitat (Escera et al., 2002; Rinne et al., 2006), i la localització del so (Roeber et al., 2003), corroborant el paper del potencial de disparitat en la detecció automàtica del canvi en general. El canvi atencional i P3a El component P3a ha estat considerat per la literatura psicofisiològica com un indicador de l’orientació de l’atenció cap el canvi (Friedman, Cycowicz, & Gaeta, 2001; Knight, 1984; Squires, Squires, & Hillyard, 1975) i la seva amplitud s’ha relacionat amb el nivell de distracció conductual. Treballs recent, però, han trobat diferències en la magnitud de la distracció en absència de la generació del component P3a (Rinne et al., 2006), i a la inversa, generació del component P3a en absència de distracció conductual (Munka & Berti, 2006; Polo et al., 2003). A més a més, mentre MMN i N1 semblen ser mecanismes que operen automàticament, P3a ha resultat altament dependent de factors moduladors superiors (top-down). Per tant, en contrast amb la més estesa interpretació de P3a com una resposta d’orientació de l’atenció, o un índex de distractibilitat conductual, la seva generació podria, més aviat, resultar un signe d’avaluació de la novetat contextual, reflectint una reconfiguració de la xarxa neuronal per tal d’actuar en conseqüència davant el canvi (Barcelo, Escera, Corral, & Perianez, 2006). La reorientació de l’atenció i RON Tant important com la flexibilitat en dirigir l’atenció cap a canvis inesperats és el retorn de la mateixa atenció cap el focus d’interès. 115 Aquest procés cognitiu s’ha estat identificat en els potencials evocats amb un component negatiu que apareix després de la generació de P3a i s’ha anomenat RON (de l’anglès, reorienting negativity; Escera et al., 2001; Schroger & Wolff, 1998b). Va ser descobert per Schroger & Wolff (1998b) quan en els seus registres una ona negativa (posteriorment batejada com a RON) apareixia només després d’estímuls discrepants que provocaven distracció, però no quan els mateixos estímuls havien de ser identificats activament o ignorats passivament pels participants. Més tard, Escera et al. (2001) va argumentar que si realment el component RON indicava l’orientació de l’atenció, la seva generació hauria de veure’s sincronitzada amb l’estímul rellevant, però no amb l’inici de l’estímul discrepant o innovador. En el seu treball, els autors van manipular l’interval temporal entre la presentació de l’estímul distractor i l’estímul rellevant i, efectivament, la generació de RON es va veure sincronitzada amb l’inici de l’estímul rellevant, independent de l’inici de l’estímul distractor. Un dels altres objectius de la present tesi (tercer estudi) serà corroborar aquesta troballa per intervals més propers i, quan l’estímul distractor i rellevant comparteixen la mateixa modalitat (auditiva). Resultats Els resultats presentats a continuació han estat extrets dels tres estudis que conformen la present tesi. En el primer dels estudis (Escera et al., 2002) l’objectiu va ser estudiar la implicació del potencial de disparitat per canvis que no fossin exclusivament en la freqüència, és a dir, per canvis també en la intensitat i la durada. En el segon dels estudis (Corral & Escera, enviat per publicació) es va explorar si en el moviment atencional provocat per estímuls innovadors era degut a un 116 moviment en l’espai, o bé, si es tractava d’un moviment dels recursos atencionals entre les modalitats sensorials implicades (visual – rellevant, auditiva –irrellevant). En el tercer dels estudis (Corral, Berti, Jacobsen, Widmann, Yago, Schroger & Escera, enviat per publicació) es va manipular l’interval de presentació del distractor i l’estímul rellevant per tal d’estudiar la seva implicació en la magnitud de la distracció, alhora que corroborar el paper del component RON en la reorientació de l’atenció. Els resultats dels tres estudis van demostrar que tant petits canvis (estímuls discrepants; primer i tercer estudi), com grans canvis (estímuls innovadors; segon estudi), respecte l’estimulació precedent, van provocar respostes més lentes. A més a més, la utilització d‘estímuls discrepants com a distractors va disminuir el nombre de respostes correctes i va augmentar el nombre d‘errors. En concret, el resultat més destacat del primer estudi va ser que tant canvis en la freqüència, com canvis en la intensitat i la durada, van generar el potencial de disparitat. La seva distribució topogràfica va ser específica per cada tipus de canvi. Pel que fa al segon estudi, la distancia espacial entre el distractor i l’estímul rellevant va resultar decisiva per la distracció conductual. En la condició en la què l’estímul distractor i l’estímul rellevant compartien la mateixa localització, o l’estímul distractor es presentava en l’hemicamp esquerra, la distracció conductual va desaparèixer. En canvi, quan el distractor es presentava en l’hemicamp dret (en anàlogues posicions que a l’hemicamp esquerra), o a través dels auriculars (situats a un 1.15 m de distància dels estímuls rellevants, com a la resta de condicions), es va observar distracció conductual. A més a més, el registre electrofisiòlogic per cada una de les condicions 117 va mostrar que l’amplitud de P3a es feia més gran conforme s’escurçava la distància entre el distractor i l’estímul rellevant. El tercer estudi va presentar patrons de distracció en funció de cada un dels dos tipus d’estímuls rellevants. Quan l’estímul rellevant va ser contestat ràpidament pels participants, canvis en la freqüència inserits en l’estímul en diferents moments van provocar respostes lentes similars i no van afectar ni el nombre de respostes correctes, ni el nombre d’errors. En canvi, en el cas en què l’estímul rellevant era respost més lentament, els mateixos canvis de freqüència inserits, van provocar diferents patrons de distracció. Quan la distància temporal entre el distractor i la característica rellevant era llarga (150 ms) les respostes van ser més ràpides però menys acurades (estil impulsiu), que quan la distància temporal era curta (50 ms), condició en la què es van presentar respostes més acurades però més lentes (estil reflexiu). Discussió Canvis inesperats en l’entorn auditiu provoquen un empitjorament del rendiment de la tasca en curs (demora en el temps de resposta). El patró d’afectació va ser diferent entre els estímuls discrepants (lleugerament diferents a l’estimulació precedent) i els estímuls innovadors (totalment diferents a l’estimulació precedent). El nombre de respostes correctes i el nombre d’errors es va veure afectat en el cas d’estímuls discrepants (estudi primer i tercer), però no en el cas d’estímuls innovadors (segon estudi). A més a més, el tipus d’estímul rellevant, així com el moment d’aparició d’un canvi de freqüència inserit en l’estímul discrepant, van resultar paràmetres influents en el patró de distracció (tercer estudi). Quan l’estímul rellevant podia ser resolt i contestat ràpidament, el canvi presentat en dos intervals 118 temporals diferents no va afectar la demora de la resposta. En canvi, quan el processament de l’estímul rellevant consumia més temps o, si més no, era contestat més lentament, canvis presentats en diferents intervals van provocar més o menys afectació a les mesures de classificació de l’estímul (nombre de respostes correctes i nombre d’errors) o la selecció/execució de la resposta (temps de reacció). Així doncs, petits canvis en l’estimulació precedent poden afectar diferents fases del processament de l’estímul rellevant en funció de la seva naturalesa i el moment de la seva presentació. D’altra banda, els estímuls innovadors no presenten cap afectació de les mesures relacionades amb la classificació de l’estímul rellevant. De fet, Parmentier et al. (2008) van suggerir que la demora provocada pels estímuls innovadors era deguda a un moviment dels recursos atencionals des de l’estímul distractor fins a l’estímul rellevant, abans de l’inici d’aquest. Els resultats del segon estudi posen de manifest que aquest moviment, més que ser degut a la diferència entre modalitats, és el resultat de moure els recursos atencionals en l’espai (de la localització on es troba l’estímul rellevant, a la localització del distractor). Tot i així, la distracció no va ser directament proporcional a la distància i el factor hemicamp va resultar important. Els distractors presentats per l’hemicamp dret, o a través dels auriculars, van provocar distracció; en canvi, els distractors que compartien la mateixa localització que l’estímul rellevant, o eren presentats per l’hemicamp esquerra en posicions homòlogues a l’hemicamp dret, no. L’activitat electrofisiològica enregistrada en els tres estudis va permetre identificar els tres components associats a la distracció conductual (potencial de disparitat, P3a i RON). El primer estudi va demostrar que el potencial de disparitat es va generar no només per canvis en la freqüència, sinó també, per canvis en la durada o la intensitat de 119 l’estimulació precedent, fet que corrobora el seu paper com a mecanisme detector automàtic del canvi. El segon estudi va observar que l’amplitud de P3a va ser sensible a la distància entre el distractor i l’estímul rellevant, motiu pel què el canvi atencional associat a P3a podria ser, de fet, un canvi en la localització espacial dels recursos atencionals. Finalment, el tercer estudi va analitzar la finestra temporal del component RON, demostrant que la seva generació depèn del moment d’afectació del distractor sobre l’estímul rellevant, és a dir, que la reorientació de l’atenció cap a la tasca depèn la fase del processament de l’estímul rellevant que afecta el distractor. Conclusions La present tesi ha investigat els indicadors electrofisiològics de la detecció i el processament de distractors auditius. A continuació es presenten les conclusions obtingudes: ƒ El tres estudis mostren que canvis inesperats en l’entorn poden provocar distracció, el què permet validar el paradigma com a eina eficaç per l’exploració de l’atenció. ƒ Els resultats obtinguts del primer i tercer estudi assenyalen que els canvis lleugerament diferents de l’estimulació precedent poden processar-se paral·lelament amb les característiques rellevants per la tasca, afectant així diferents fases del processament de l’estímul rellevant en funció del moment d’aparició del distractor. Contràriament, quan el canvi resulta totalment diferent a l’estimulació precedent, els recursos atencionals es dirigeixen completament a l’avaluació de l’estímul innovador, demorant el 120 processament de l’estímul rellevant, però no afectant cap de les seves fases. ƒ La manipulació de la localització del distractor en el segon estudi va resultar un factor important per la distracció. Quan el distractor es presenta en la mateixa posició que l’estímul rellevant, o per l’hemicamp esquerra del participant, la distracció va quedar abolida. En canvi, distractors presentats per l’hemicamp dret, o pels auriculars, van mostrar distracció. ƒ El potencial de disparitat registrat en el primer estudi es va presentar davant canvis en la freqüència, la durada i la intensitat, resultats que corroboren el seu paper com a mecanisme automàtic en la detecció del canvi. ƒ En el segon estudi, el canvi atencional associat a P3a va ser sensible a la distància espacial entre el distractor i l’estímul rellevant, suggerint que el canvi d’atenció de l’estímul rellevant a l’estímul distractor, en realitat, impliqui un canvi dels recursos atencionals en l’espai. ƒ El procés cognitiu de reorientació de l’atenció cap a la tasca principal, després d’una distracció momentània, analitzat en el tercer estudi, va resultar dependent de la fase del processament de l’estímul rellevant afectat pel distractor. 121 Referències Alho, K. (1995). Cerebral generators of mismatch negativity (MMN) and its magnetic counterpart (MMNm) elicited by sound changes. Ear and Hearing, 16(1), 38-51. Alho, K., Escera, C., Diaz, R., Yago, E., & Serra, J. M. (1997). Effects of involuntary auditory attention on visual task performance and brain activity. Neuroreport, 8(15), 3233-3237. Alho, K., Winkler, I., Escera, C., Huotilainen, M., Virtanen, J., Jaaskelainen, I. P., et al. (1998). Processing of novel sounds and frequency changes in the human auditory cortex: Magnetoencephalographic recordings. Psychophysiology, 35(2), 211-224. Barcelo, F., Escera, C., Corral, M. J., & Perianez, J. A. (2006). Task switching and novelty processing activate a common neural network for cognitive control. Journal of Cognitive Neuroscience, 18(10), 1734-1748. Berti, S., Roeber, U., & Schroger, E. (2004). Bottom-up influences on working memory: Behavioral and electrophysiological distraction varies with distractor strength. Experimental Psychology, 51(4), 249-257. Corral, M.J. & Escera, C. Effects of sound location on visual task performance and electrophysiological measures of distraction. Neuroreport, en premsa. Corral, M.J., Berti, S., Jacobsen, T., Widmann, A., Yago, E., Schroger, E., & Escera, C. (enviat per publicació). Distraction effects with different distractor-to-target intervals: a combined behavioral and event-related brain potential study. Escera, C., Alho, K., Schroger, E., & Winkler, I. (2000). Involuntary attention and distractibility as evaluated with event-related brain potentials. Audiology & Neuro-Otology, 5(3-4), 151-166. Escera, C., Alho, K., Winkler, I., & Naatanen, R. (1998). Neural mechanisms of involuntary attention to acoustic novelty and change. Journal of Cognitive Neuroscience, 10(5), 590-604. 122 Escera, C., Corral, M. J., & Yago, E. (2002). An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, duration and intensity changes. Brain Research. Cognitive Brain Research, 14(3), 325-332. Escera, C., Yago, E., & Alho, K. (2001). Electrical responses reveal the temporal dynamics of brain events during involuntary attention switching. The European Journal of Neuroscience, 14(5), 877-883. Escera, C., Yago, E., Corral, M. J., Corbera, S., & Nunez, M. I. (2003). Attention capture by auditory significant stimuli: Semantic analysis follows attention switching. The European Journal of Neuroscience, 18(8), 2408-2412. Deouell, L.Y. (2007). The frontal generator of the mismatch negativity revisited. Journal of Psychophysiology, 21(3-4), 188-203. Friedman, D., Cycowicz, Y. M., & Gaeta, H. (2001). The novelty P3: An event-related brain potential (ERP) sign of the brain's evaluation of novelty. Neuroscience and Biobehavioral Reviews, 25(4), 355-373. Giard, M. H., Perrin, F., Echallier, J. F., Thevenet, M., Froment, J. C., & Pernier, J. (1994). Dissociation of temporal and frontal components in the human auditory N1 wave: A scalp current density and dipole model analysis. Electroencephalography and Clinical Neurophysiology, 92(3), 238-252. Jaaskelainen, I. P., Schroger, E., & Naatanen, R. (1999). Electrophysiological indices of acute effects of ethanol on involuntary attention shifting. Psychopharmacology, 141(1), 1621. Jacobsen, T., & Schroger, E. (2001). Is there pre-attentive memorybased comparison of pitch? Psychophysiology, 38(4), 723-727. Knight, R. T. (1984). Decreased response to novel stimuli after prefrontal lesions in man. Electroencephalography and Clinical Neurophysiology, 59(1), 9-20. Munka, L., & Berti, S. (2006). Examining task-dependencies of different attentional processes as reflected in the P3a and reorienting negativity components of the human event-related brain potential. Neuroscience Letters, 396(3), 177-181. 123 Naatanen, R. (1990). The role of attention in auditory information processing as revealed by event-related potentials and other brain measures of cognitive function. Behavioral and Brain Sciences, 13, 201-288. Naatanen, R., (2007). The mismatch negativity (MMN): Where is the big fish? Journal of Psychophysiology, 21(3-4), 133-137. Naatanen, R., & Picton, T. (1987). The N1 wave of the human electric and magnetic response to sound: A review and an analysis of the component structure. Psychophysiology, 24(4), 375-425. Parmentier, F. B., Elford, G., Escera, C., Andres, P., & San Miguel, I. (2008). The cognitive locus of distraction by acoustic novelty in the cross-modal oddball task. Cognition, 106(1), 408-432. Polo, M. D., Escera, C., Yago, E., Alho, K., Gual, A., & Grau, C. (2003). Electrophysiological evidence of abnormal activation of the cerebral network of involuntary attention in alcoholism. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 114(1), 134-146. Rinne, T., Sarkka, A., Degerman, A., Schroger, E., & Alho, K. (2006). Two separate mechanisms underlie auditory change detection and involuntary control of attention. Brain Research, 1077(1), 135-143. Roeber, U., Berti, S., & Schroger, E. (2003). Auditory distraction with different presentation rates: An event-related potential and behavioral study. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 114(2), 341349. Roeber, U., Widmann, A., & Schroger, E. (2003). Auditory distraction by duration and location deviants: A behavioral and event-related potential study. Brain Research.Cognitive Brain Research, 17(2), 347-357. Schroger, E. (1996). Neural mechanism for involuntary attention shifts to changes in auditory stimulation. Journal of Cognitive Neuroscience, 8, 527-539. Schroger, E. (2007). Mismatch negativity: a microphone into auditory memory. Journal of Psychophysiology, 21(3-4), 138-146. 124 Schroger, E., Giard, M. H., & Wolff, C. (2000). Auditory distraction: Event-related potential and behavioral indices. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 111(8), 1450-1460. Schroger, E., & Wolff, C. (1998a). Behavioral and electrophysiological effects of task-irrelevant sound change: A new distraction paradigm. Brain Research.Cognitive Brain Research, 7(1), 71-87. Schroger, E., & Wolff, C. (1998b). Attentional orienting and reorienting is indicated by human event-related brain potentials. Neuroreport, 9(15), 3355-3358. Squires, N. K., Squires, K. C., & Hillyard, S. A. (1975). Two varieties of long-latency positive waves evoked by unpredictable auditory stimuli in man. Electroencephalography and Clinical Neurophysiology, 38(4), 387-401. Sussman, E., Winkler, I., & Schroger, E. (2003). Top-down control over involuntary attention switching in the auditory modality. Psychonomic Bulletin & Review, 10(3), 630-637. Winkler, I. (2007). Interpreting the mismatch negativity (MMN). Journal of Psychophysiology, 21(3-4), 147-163. Yago, E., Corral, M. J., & Escera, C. (2001). Activation of brain mechanisms of attention switching as a function of auditory frequency change. Neuroreport, 12(18), 4093-4097. 125 Theoretical reviews Escera, C. & Corral, M.J. (2007). Role of mismatch negativity and noveltyP3 in involuntary auditory attention. Journal of Psychophysiology, 21 (3-4), 251-264. Escera, C. & Corral, M.J. (2003). The distraction potential (DP), an electrophysiological tracer of involuntary attention control and its dysfunction. In I. Reinvang, M.W. Greenlee & M. Herrmann (Eds.), The cognitive neuroscience of individual differences (63-76). Oldenburg: Bibliotheks-und Informationsystem der Universität Oldenburg.