Human Visual System Automatically Encodes Sequential Regularities of Discrete Events
Total Page:16
File Type:pdf, Size:1020Kb
Human Visual System Automatically Encodes Sequential Regularities of Discrete Events Motohiro Kimura1,2,3, Erich Schröger3, István Czigler4, and Hideki Ohira1 Downloaded from http://mitprc.silverchair.com/jocn/article-pdf/22/6/1124/1769676/jocn.2009.21299.pdf by guest on 18 May 2021 Abstract ■ For our adaptive behavior in a dynamically changing environ- sisting of infrequent deviant and frequent standard stimuli, and ment, an essential task of the brain is to automatically encode se- tested whether the underlying memory representation of visual quential regularities inherent in the environment into a memory MMN generation contains only a sensory memory trace of stan- representation. Recent studies in neuroscience have suggested dard stimuli (trace-mismatch hypothesis) or whether it also that sequential regularities embedded in discrete sensory events contains sequential regularities extracted from the repetitive are automatically encoded into a memory representation at the standard sequence (regularity-violation hypothesis). The results level of the sensory system. This notion is largely supported showed that visual MMN was elicited by first deviant (deviant by evidence from investigations using auditory mismatch negativ- stimuli following at least one standard stimulus), second de- ity (auditory MMN), an event-related brain potential (ERP) corre- viant (deviant stimuli immediately following first deviant), and late of an automatic memory-mismatch process in the auditory first standard (standard stimuli immediately following first devi- sensory system. However, it is still largely unclear whether or ant), but not by second standard (standard stimuli immediately not this notion can be generalized to other sensory modalities. following first standard). These results are consistent with the The purpose of the present study was to investigate the contribu- regularity-violation hypothesis, suggesting that the visual sen- tion of the visual sensory system to the automatic encoding of sory system automatically encodes sequential regularities. In sequential regularities using visual mismatch negativity (visual combination with a wide range of auditory MMN studies, the MMN), an ERP correlate of an automatic memory-mismatch present study highlights the critical role of sensory systems in process in the visual sensory system. To this end, we conducted automatically encoding sequential regularities when modeling a sequential analysis of visual MMN in an oddball sequence con- the world. ■ INTRODUCTION tant role in the automatic encoding of a large amount of An external environment is dynamically changing. For our sensory information. However, its contents are rather adaptive behavior in such a changing environment, an literal, in the sense that it is involved in encoding static essential task of the brain is to automatically encode what information rather than dynamic information, such as is regular in a sequence of sensory events into a memory sequential regularities embedded in a sequence of sensory representation, even when the sequence is irrelevant for events. the current task, as such a memory representation forms In contrast to this traditional notion, recent neuro- the basis of predictions for future events, which enables science studies have provided evidence which supports us to minimize the processing resources for redundant the notion that not only static information but also dynamic events and to maximize the processing resources for novel information, such as sequential regularities, is, in fact, events (e.g., Sokolov, 1963). automatically encoded into a memory representation at When a sequence of sensory events is irrelevant for the level of the sensory system. This evidence is largely the current task, an automatically formed memory rep- obtained from investigations using auditory mismatch resentation has traditionally been understood in terms negativity (auditory MMN; Näätänen, Gaillard, & Mäntysalo, of a sensory store, which was originally described as a 1978; for reviews, see e.g., Näätänen, Paavilainen, Rinne, & memory register in the multistore model proposed by Alho, 2007; Näätänen, Tervaniemi, Sussman, Paavilainen, & Atkinson and Shiffrin (1968). According to this model, Winkler, 2001; Näätänen, 1990, 1992) event-related brain the sensory store is sensory-specific and plays an impor- potential (ERP). A series of auditory MMN studies have shed new light on the involvement of the auditory sen- sory system in higher cognitive processes than previously 1Nagoya University, Nagoya, Japan, 2Japan Society for the Promo- thought (e.g., “primitive intelligence”; Näätänen et al., tion of Science, Japan, 3University of Leipzig, Leipzig, Germany, 2001) and provide important insight into recent theories 4Hungarian Academy of Sciences, Budapest, Hungary on the brain which state that, rather than passively waiting © 2009 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 22:6, pp. 1124–1139 Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/jocn.2009.21299 by guest on 26 September 2021 to be activated by sensory events, the brain is constantly & Balázs, 2005). Visual MMN has generators in visual areas predicting future events (e.g., “proactive brain,” Bar, 2007; (Amenedo et al., 2007; Yucel, McCarthy, & Belger, 2007; “predictive coding,” Friston, 2005; “predictive modeling,” Czigler et al., 2004; Pazo-Alvarez, Amenedo, Lorenzo-López, Denham & Winkler, 2006). et al., 2004) and is largely unaffected by several attentional Although auditory MMN studies have provided evidence manipulations (Pazo-Alvarez, Amenedo, & Cadaveira, 2004; which supports this new notion, it is still largely unclear Heslenfeld, 2003), which suggests that visual MMN gen- whether this notion can be generalized to other sensory eration is an automatic brain process in the visual sensory modalities. The aim of the present study was to investigate system. the contribution of the visual sensory system to the auto- matic encoding of sequential regularities using a visual Downloaded from http://mitprc.silverchair.com/jocn/article-pdf/22/6/1124/1769676/jocn.2009.21299.pdf by guest on 18 May 2021 MMN and the Involvement of analogue of MMN, visual mismatch negativity (visual MMN; Memory Representation e.g., Kimura, Katayama, Ohira, & Schröger, 2009; Czigler, Balázs, & Winkler, 2002; for reviews, see Czigler, 2007; With regard to the processes that underlie MMN genera- Pazo-Alvarez, Cadaveira, & Amenedo, 2003). tion, two contrasting hypotheses have been proposed: the memory-mismatch hypothesis and the refractoriness hypothesis. The memory-mismatch hypothesis assumes Mismatch Negativity that MMN is an ERP correlate of a memory-mismatch Auditory MMN is a negative-going ERP component that process and is elicited when a current stimulus input mis- peaks at around 150–250 msec after the onset of deviant matches the memory representation formed by the pre- stimulation with a fronto-central scalp distribution, and ceding stimulus sequence (Näätänen, 1990, 1992). In has been typically observed in response to infrequent “de- contrast, the refractoriness hypothesis assumes that MMN viant” compared to frequent “standard” stimuli in the is not an ERP correlate of a memory-mismatch process auditory oddball sequence. For example, it is elicited by in- and fully or partly reflects a lower refractoriness level (or frequent deviations in several stimulus dimensions such a lower adaptation level) of afferent N1 neurons that spe- as frequency (Näätänen et al., 1978), intensity (Näätänen cifically respond to a feature value of infrequent deviant et al., 1978), duration (Näätänen, Paavilainen, & Reinikainen, stimuli compared to those that specifically respond to a fea- 1989), and location (Paavilainen, Karlsson, Reinikainen, & ture value of frequent standard stimuli. Thus, MMN may Näätänen, 1989) (for reviews, see e.g., Näätänen, 1990, result from the small amplitude of N1 elicited by frequent 1992). Auditory MMN has generators in the auditory (and standard stimuli and the large amplitude of N1 elicited by probably frontal) areas (for a review, see Deouell, 2007) infrequent deviant stimuli ( Jääskeläinen et al., 2004). and is largely unaffected by various attentional manipula- Concerning this issue, several studies have shown that tions (for a review, see Sussman, 2007), which suggests that although it is often overlapped by N1, reflecting the refrac- auditory MMN generation is an automatic brain process in tory effect, MMN has several characteristics that are differ- the auditory sensory system. ent from those of N1 and cannot be explained in terms of It has long been unclear whether or not MMN occurs in the refractoriness hypothesis. For example, unlike audi- other sensory modalities. Over the past few decades, how- tory N1, auditory MMN is elicited even by infrequent in- ever, several studies have tried to identify a visual analogue tensity decrements (Woldorff, Hackley, & Hillyard, 1991; of MMN and have reported that infrequent deviant compared Näätänen, Paavilainen, Alho, Reinikainen, & Sams, 1989), to frequent standard stimuli in the visual oddball sequence infrequent duration decrements (Kaukoranta, Sams, Hari, elicited a posterior negativity that peaked at around 150– Hämäläinen, & Näätänen, 1989; Näätänen, Paavilainen, & 250 msec after stimulus onset and is usually called visual Reinikainen, 1989), and sudden stimulus omission (Yabe MMN. For example, it is elicited by infrequent deviations in et al., 1998; Yabe, Tervaniemi, Reinikainen, & Näätänen,