Repetition and the brain: neural models of stimulus-specific effects

https://doi.org/10.1016/j.tics.2005.11.006Get rights and content

One of the most robust experience-related cortical dynamics is reduced neural activity when stimuli are repeated. This reduction has been linked to performance improvements due to repetition and also used to probe functional characteristics of neural populations. However, the underlying neural mechanisms are as yet unknown. Here, we consider three models that have been proposed to account for repetition-related reductions in neural activity, and evaluate them in terms of their ability to account for the main properties of this phenomenon as measured with single-cell recordings and neuroimaging techniques. We also discuss future directions for distinguishing between these models, which will be important for understanding the neural consequences of repetition and for interpreting repetition-related effects in neuroimaging data.

Introduction

When stimuli are repeated, neural activity is usually reduced. This neural repetition effect has been reported at multiple spatial scales, from the level of individual cortical neurons in monkeys 1, 2, 3 to the level of hemodynamic changes (measuring the pooled activation of millions of neurons) in humans using functional magnetic resonance imaging (e.g. fMRI 4, 5, 6, 7, 8, 9, 10). Repetition-related reductions also occur at multiple temporal scales, both in their longevity – from milliseconds [3] to minutes [9] and days [11] – and in the latency of their expression 12, 13. The phenomenon also occurs in multiple brain regions, and across an impressively large number of experimental conditions.

This stimulus-specific reduction in neural activity has been referred to as adaptation 14, 15, 16, mnemonic filtering [17], repetition suppression [18], decremental responses [19] and neural priming [20]. We will use ‘repetition suppression’ (RS) to refer to decreased neural responses following stimulus repetition. As will be apparent below, it is important to distinguish the scale at which RS arises, because the correspondence across scales (e.g. neural firing and hemodynamic responses) might not be simple.

Interest in repetition effects has recently intensified, for two main reasons. First, repetition effects have proved useful for inferring the nature of representations across different stages of a processing stream. This approach has been used behaviorally (e.g. using visual aftereffects to infer the nature of orientation tuning [21] or face representation 22, 23), with single-cell recording [24], and more recently has become popular with fMRI, particularly given claims that it enables improved spatial resolution [7] (Box 1). The second reason for heightened interest is the possibility that RS might be the neural correlate of priming 25, 26, 27. Priming refers to improved processing of a repeated stimulus according to some behavioral measure (e.g. greater accuracy in identifying the stimulus, or faster response times to make a decision about it), and often occurs under the same experimental conditions as RS. Nevertheless, it is important to note that, under certain conditions, priming can be associated with increased activity, rather than reduction (for discussion of repetition enhancement effects and changes in frontal cortices that might contribute to priming, see 27, 28).

The purpose of this review is to consider several kinds of neural models that have been proposed to account for repetition suppression (RS). We focus primarily on studies using visually presented objects and their effects on the ventral object processing stream, to maximize overlap between monkey and human studies. We evaluate the neural models in terms of their ability to account for the main properties of RS as measured with single-cell recordings, fMRI and electroencephalogram/magnetoencephalogram (EEG/MEG), discuss implications of these models for interpreting experimental results, and propose directions for distinguishing between the models.

Section snippets

Repetition suppression as measured with single-cell recording

Stimulus-specific repetition-related reductions in firing rates have been found in physiological recordings of neurons in macaque inferior temporal (IT) cortex 14, 15, 17, 18, 19 (Figure 1). These repetition effects have been reported for awake behaving animals performing various visual tasks (e.g. match to sample [17], recognition memory 19, 29), as well as in anesthetized animals [30], and occur for both behaviorally relevant and irrelevant stimuli [2]. RS effects are stimulus specific in the

Repetition suppression as measured with fMRI

The basic phenomenon of RS measured with fMRI, also referred to as fMRI-adaptation, has been replicated many times in ventral temporal cortex (see review [32]) as well other areas such as medial temporal [6] and frontal cortex 33, 34, 35. Many adaptation paradigms have been used to measure RS, including multiple repetitions of the same stimulus without intervening items [7] (Figure 2a: block design), or after a single presentation with either no [36] or many [37] intervening items (Figure 2a:

Repetition effects as measured with EEG/MEG

Repetition effects have also been studied by measuring changes in the electrical (EEG) or magnetic (MEG) field, usually recorded above the scalp. These effects reflect changes in the amplitude and/or synchrony of local field potentials (LFPs) caused by transmembrane currents in large numbers of neurons. Most EEG studies examine event-related potentials (ERPs), which reflect changes in electrical potential during the few hundred milliseconds following stimulus onset, averaged across trials. The

Relating different types of data

Relating the data recorded from single cells by electrodes (firing rates, multi-unit activity: MUA, or LFP) to the data recorded by fMRI (BOLD) and by EEG/MEG (synchronous changes in LFPs over many neurons) is non-trivial. In general terms, increases in MUA correlate with increases in LFP, and increases in LFP correlate with increases in BOLD [58] and presumably EEG/MEG power. However, the special constraints of each technique give rise to situations where their measures might dissociate. For

Models for repetition suppression

There are multiple potential neural causes of the RS measured with single-cell recording, fMRI or EEG/MEG. We now describe three models that have been sketched previously in the literature (Figure 4): (i) the Fatigue model, whereby the amplitude of firing of stimulus-responsive neurons decreases 2, 16; (ii) the Sharpening model, whereby fewer neurons respond 1, 18, 26; and (iii) the Facilitation model, whereby the latency [60] and/or duration of neural activity is shortened 3, 27. An important

Distinguishing the neural models

There are three main directions in which these models can be distinguished: (i) examining the relationship between RS and stimulus selectivity; (ii) examining the effect of repetition on the tuning of cortical responses along a stimulus dimension; and (iii) examining the temporal window in which processing occurs for new and repeated stimuli.

Conclusions

Adaptation paradigms have become increasingly popular for examining the processing characteristics of different cortical regions of the brain. Clearly, progress has been made using RS to infer the nature of representations in different cortical regions, or as a marker for increased processing efficiency, without a complete understanding of its neural basis. Nonetheless, it is our belief that the specific neural mechanisms matter, because interpretation and design of experiments depend on the

Acknowledgements

The authors have contributed equally to this manuscript. We would like to thank Steve Gotts for many helpful conversations, Rory Sayres and Sven Heinrich for contributing to Figure 2, and Thomas Gruber for contributing Figure 3. We would also like to thank Dave Andresen, Sharon Gilai-Dotan, Sven Heinrich, David Leopold, David McMahon, Rafael Malach, Mortimer Mishkin, Carl Olson, Leslie Ungerleider, Anthony Wagner, and Nathaniel Witthoff for their insightful comments on the manuscript. Finally,

References (94)

  • J.Z. Xiang et al.

    Differential neuronal encoding of novelty, familiarity and recency in regions of the anterior temporal lobe

    Neuropharmacology

    (1998)
  • G. Avidan

    Analysis of the neuronal selectivity underlying low fMRI signals

    Curr. Biol.

    (2002)
  • T.W. James

    Differential effects of viewpoint on object-driven activation in dorsal and ventral streams

    Neuron

    (2002)
  • Z. Kourtzi

    Integration of local features into global shapes: monkey and human fMRI studies

    Neuron

    (2003)
  • G.M. Doniger

    Visual perceptual learning in human object recognition areas: a repetition priming study using high-density electrical mapping

    Neuroimage

    (2001)
  • M. Eimer

    Event-related brain potentials distinguish processing stages involved in face perception and recognition

    Clin. Neurophysiol.

    (2000)
  • M.J. Brookes

    GLM-beamformer method demonstrates stationary field, alpha ERD and gamma

    Neuroimage

    (2005)
  • R.N. Henson

    Detecting latency differences in event-related BOLD responses: application to words versus nonwords and initial versus repeated face presentations

    Neuroimage

    (2002)
  • W. Koutstaal

    Perceptual specificity in visual object priming: functional magnetic resonance imaging evidence for a laterality difference in fusiform cortex

    Neuropsychologia

    (2001)
  • S.A. Engel

    Adaptation of oriented and unoriented color-selective neurons in human visual areas

    Neuron

    (2005)
  • J.C. Culham

    Independent aftereffects of attention and motion

    Neuron

    (2000)
  • K. Grill-Spector

    The neural basis of object perception

    Curr. Opin. Neurobiol.

    (2003)
  • L. Li

    The representation of stimulus familiarity in anterior inferior temporal cortex

    J. Neurophysiol.

    (1993)
  • E.K. Miller et al.

    Parallel neuronal mechanisms for short-term memory

    Science

    (1994)
  • S. Sobotka et al.

    Mnemonic responses of single units recorded from monkey inferotemporal cortex, accessed via transcommissural versus direct pathways: a dissociation between unit activity and behavior

    J. Neurosci.

    (1996)
  • J.B. Demb

    Semantic encoding and retrieval in the left inferior prefrontal cortex: a functional MRI study of task difficulty and process specificity

    J. Neurosci.

    (1995)
  • R.L. Buckner

    Functional anatomical studies of explicit and implicit memory retrieval tasks

    J. Neurosci.

    (1995)
  • C.E. Stern

    The hippocampal formation participates in novel picture encoding: evidence from functional magnetic resonance imaging

    Proc. Natl. Acad. Sci. U. S. A.

    (1996)
  • Y. Jiang

    Complementary neural mechanisms for tracking items in human working memory

    Science

    (2000)
  • R. Henson

    Neuroimaging evidence for dissociable forms of repetition priming

    Science

    (2000)
  • L. Naccache et al.

    The priming method: imaging unconscious repetition priming reveals an abstract representation of number in the parietal lobes

    Cereb. Cortex

    (2001)
  • M. van Turennout

    Long-lasting cortical plasticity in the object naming system

    Nat. Neurosci.

    (2000)
  • E.K. Miller

    Activity of neurons in anterior inferior temporal cortex during a short-term memory task

    J. Neurosci.

    (1993)
  • R. Desimone

    Neural mechanisms for visual memory and their role in attention

    Proc. Natl. Acad. Sci. U. S. A.

    (1996)
  • L. Maccotta et al.

    Evidence for neural effects of repetition that directly correlate with behavioral priming

    J. Cogn. Neurosci.

    (2004)
  • N. Graham

    Visual Pattern Analyzers

    (1989)
  • M.A. Webster

    Adaptation to natural facial categories

    Nature

    (2004)
  • D.A. Leopold

    Prototype-referenced shape encoding revealed by high-level aftereffects

    Nat. Neurosci.

    (2001)
  • A. Lueschow

    Inferior temporal mechanisms for invariant object recognition

    Cereb. Cortex

    (1994)
  • D.L. Schacter

    Specificity of priming: a cognitive neuroscience perspective

    Nat. Rev. Neurosci.

    (2004)
  • S. Sobotka et al.

    Investigation of long-term recognition and association memory in unit responses from inferotemporal cortex

    Exp. Brain Res.

    (1993)
  • E.K. Miller et al.

    Scopolamine affects short-term memory but not inferior temporal neurons

    Neuroreport

    (1993)
  • Z. Kourtzi et al.

    fMRI adaptation: a tool for studying visual representations in the primate brain

  • A.D. Wagner

    Dissociations between familiarity processes in explicit recognition and implicit perceptual memory

    J. Exp. Psychol. Learn. Mem. Cogn.

    (1997)
  • R.L. Buckner et al.

    Functional neuroimaging studies of encoding, priming, and explicit memory retrieval

    Proc. Natl. Acad. Sci. U. S. A.

    (1998)
  • M. van Turennout

    Modulation of neural activity during object naming: effects of time and practice

    Cereb. Cortex

    (2003)
  • Z. Kourtzi et al.

    Cortical regions involved in perceiving object shape

    J. Neurosci.

    (2000)
  • Cited by (1813)

    View all citing articles on Scopus
    View full text