Trends in Cognitive Sciences
Perirhinal cortical contributions to object perception
Introduction
Substantial damage to the medial temporal lobe (MTL), which comprises the hippocampus and underlying perirhinal, entorhinal and parahippocampal cortices (see Figure 1), results in dense amnesia. For example, patients such as H.M., who received bilateral anterior medial temporal lobectomy as a treatment for intractable epilepsy, become profoundly amnesic [1]. Similarly, macaque monkeys with lesions to the hippocampus and amygdala are severely impaired at recognition memory [2]. Patients with more selective MTL damage, centred upon the hippocampus, also show memory deficits 3, 4, 5, 6. These observations have led to the formulation of the influential theory that the primate hippocampus lies at the centre of a MTL-based declarative memory system [7]. According to this theory, the MTL is essential for the consolidation of declarative memories; such memories for facts and experiences are thought to become independent from the hippocampus with time, as they are consolidated into long-term memories that are dependent upon the neocortex. MTL memory system theory in its canonical form asserts that the MTL is exclusively involved in memory. According to this view, visual perception and visual memory are entirely dissociable cognitive processes with visual perception proposed to depend on inferotemporal cortex and more posterior cortical regions in the occipital lobe.
However, in the past decade, data has emerged in macaque studies that directly challenges MTL memory system theory 8, 9, 10, 11, 12, 13, 14, 15; indeed six years ago, a TICS review [16] of these initial studies concluded that the perirhinal cortex (PRh) in the MTL might contribute to both object memory and perception. This provoked ardent defence of the MTL memory system theory with some authors rigorously maintaining that the PRh does not contribute to perception 17, 18, 19, 20, 21. Determining the role of the PRh has become the crux of the argument as researchers debate whether or not it contributes to perception. Here we review the latest evidence for and against the MTL memory system theory, explaining the rationale behind the polarized claims and counter-claims that pervade this literature. Our opinion, though, is that the PRh does contribute to perception. The implications of this view, if substantiated, are that memory is not as dissociable from other aspects of cognition as standard MTL memory system theory contends, and accordingly that the majority of psychologists and neuroscientists might have held a fundamentally flawed view of how memory is organized in the brain.
Section snippets
Evaluating the evidence
Readers new to this debate might expect that this issue could be settled one way or the other with relative ease. For example, if one or more perceptual tasks could be shown to be PRh-dependent then the PRh cannot contribute exclusively to memory. That the controversy continues unabated can largely be attributed to two factors. First, as argued by Hampton [22], several tasks used to support claims that the PRh does contribute to perception confound mnemonic with perceptual processes. However,
Evidence from recognition memory
The severe recognition memory impairments that follow amygdala and hippocampal lesions in macaques [2] are now largely attributed to the damage sustained to underlying PRh 23, 24, 25. Successful recognition memory performance (see Figure 2a) requires intact perceptual discrimination and familiarity judgements, although it has been argued that the presence of delay-dependent deficits might indicate a mnemonic impairment whereas poor performance in zero or very short delay conditions might point
Evidence from concurrent discrimination learning (CDL)
Another task, concurrent discrimination learning (CDL; see Figure 2b), has also been widely used to probe the role of MTL structures. As early studies failed to find CDL deficits after MTL lesions, CDL was originally viewed as a habit-learning task dependent upon the striatum as opposed to a MTL-dependent memory task. This was despite the fact that, unlike recognition memory, CDL cannot be solved on the basis of familiarity judgements alone; CDL also requires associative memory for
The effect of set-size
However, Hampton [22] argues that because no more errors-per-problem accrued when larger set sizes were investigated [9], that there was no significant correlation between magnitude of impairment on CDL and set-size after our PRh lesions. This does not alter the fact that so many studies have failed to find deficits with small set sizes (see 30, 31 for a review), which contrasts with the significant deficits observed when large set sizes are used. We predicted that large set-sizes would reveal
Levels of representation
To summarize so far, our thesis is that TE and PRh are distinguished, not by any differential contribution to perception or memory, but according to the different levels of stimulus representation that they support (see Box 2). According to this view, TE contributes to both memory and perception of simple features by virtue of its specialization for representing simple features. Similarly, PRh is only recruited in tasks that place sufficient demands upon memory or perception for objects because
De-confounding memory and perception
Because of the confounding of memory and perception, neither recognition memory nor CDL provides unambiguous evidence of whether PRh contributes to memory alone or also to perception. However, macaques with rhinal (combined PRh and entorhinal) lesions were impaired at recognition memory when trial-unique stimuli but not when very small stimulus sets were used [12]. With such small set-sizes, rewarded test items on one trial inevitably occur as unrewarded test items in other trials thereby
Generalization to new views
In an attempt to modulate the perceptual demands of CDL, while keeping the memory demands unchanged, we conducted an experiment in which animals were required to discriminate familiar stimuli presented in similar but previously unseen views [11]. Although PRh-lesioned animals were impaired, it was argued that our errors-to-criterion measure used (animals continued daily testing on the same set of new views until they attained criterion) might have confounded re-learning with stimulus
The oddity task
To assess perceptual deficits after PRh lesions more directly, we developed a simultaneous visual discrimination task for macaques called ‘Oddity’ [8]. Figure 3 illustrates example oddity problems and details the results from the different stages of the study. We found that PRh lesions impaired macaques' abilities to discriminate between objects but not their abilities to make difficult discriminations between simple features. Furthermore, their object discrimination deficits were specific to
The morph task
Bussey and colleagues pioneered a different approach to investigating PRh contributions to perception. Their hypothesis [16] is that PRh is necessary for representing complex conjunctions of features; furthermore, their connectionist modelling accurately predicted that macaques with PRh lesions would be impaired at acquiring discriminations with high but not low degrees of feature ambiguity [15]. Their model also predicts that discriminating individual problems with high feature overlap will be
Conclusions
The convergence of findings across different species and across different paradigms leads us to conclude that the primate PRh contributes to perception and substantiates our challenge to the popular theory that the MTL houses a memory system exclusively. Freed from the constraints of the MTL memory system hypothesis, the aim of future research should be to investigate what contribution the MTL makes to a broader array of cognitive processes such as spatial perception 38, 39, 40 and temporal
References (60)
- et al.
Perceptual–mnemonic functions of the perirhinal cortex
Trends Cogn Sci.
(1999) Spared short-term-memory in monkeys following medial temporal- lobe lesions is not yet established
A reply. Behav. Brain Res.
(1993)Memory decays at the same rate in macaques with and without brain lesions when expressed in d' or arcsine terms
Behav. Brain Res.
(1991)- et al.
Perirhinal cortex ablation impairs configural learning and paired-associate learning equally
Neuropsychologia
(1998) Perceptual and mnemonic matching-to-sample in humans: contributions of the hippocampus, perirhinal and other medial temporal lobe cortices
Cortex
(2000)Perceptual deficits in amnesia: challenging the medial temporal lobe ‘mnemonic’ view
Neuropsychologia
(2005)What is a memory system? Horel's critique revisited
Behav. Brain Res.
(2001)Dense amnesia in the monkey after transection of fornix, amygdala and anterior temporal stem
Neuropsychologia
(2001)- et al.
Crossed unilateral lesions of the medial forebrain bundle and either inferior temporal or frontal cortex impair object-reward association learning in Rhesus monkeys
Neuropsychologia
(2001) The anatomy, physiology and functions of the perirhinal cortex
Curr. Opin. Neurobiol.
(1996)
What's new with the amnesic patient H.M?
Nat. Rev. Neurosci.
Memory in monkeys severely impaired by combined but not by separate removal of amygdala and hippocampus
Nature
Retrograde amnesia for facts and events: findings from four new cases
J. Neurosci.
Three cases of enduring memory impairment after bilateral damage limited to the hippocampal formation
J. Neurosci.
Human amnesia and the medial temporal region: enduring memory impairment following a bilateral lesion limited to field CA1 of the hippocampus
J. Neurosci.
Differential effects of early hippocampal pathology on episodic and semantic memory
Science
The medial temporal lobe memory system
Science
Selective perceptual impairments after perirhinal cortex ablation
J. Neurosci.
Impairment of visual object-discrimination learning after perirhinal cortex ablation
Behav. Neurosci.
Learning and transfer of object–reward associations and the role of the perirhinal cortex
Behav. Neurosci.
Perirhinal cortex ablation impairs visual object identification
J. Neurosci.
Preserved recognition memory for small sets, and impaired stimulus identification for large sets, following rhinal cortex ablations in monkeys
Eur. J. Neurosci.
The organisation of visual object representations: a connectionist model of effects of lesions in perirhinal cortex
Eur. J. Neurosci.
Impairments in visual discrimination after perirhinal cortex lesions: testing ‘declarative’ vs. ‘perceptual-mnemonic’ views of perirhinal cortex function
Eur. J. Neurosci.
Perirhinal cortex resolves feature ambiguity in complex visual discriminations
Eur. J. Neurosci.
Dissociation between the effects of damage to perirhinal cortex and area TE
Learn. Mem.
Perception and recognition memory in monkeys following lesions of area TE and perirhinal cortex
Learn. Mem.
The human perirhinal cortex and recognition memory
Hippocampus
A reexamination of the concurrent discrimination learning task: the importance of anterior inferotemporal cortex, area TE
Behav. Neurosci.
Intact visual discrimination of complex and feature-ambiguous stimuli in the absence of perirhinal cortex
Learn. Mem.
Cited by (133)
Representation of human spatial navigation responding to input spatial information and output navigational strategies: An ALE meta-analysis
2019, Neuroscience and Biobehavioral ReviewsRhythmic Pruning of Perceptual Noise for Object Representation in the Hippocampus and Perirhinal Cortex in Rats
2019, Cell ReportsCitation Excerpt :Moreover, neurons that responded to nonspecific combinations of objects were more abundant in the hippocampus than in the PER. These results suggest that PER neurons outperform hippocampal cells in generating stimulus-selective repetition effects (Ahn and Lee, 2017; Buckley et al., 2001; Buckley and Gaffan, 2006; Bussey et al., 2006). In contrast, the hippocampus may be more attuned to forming arbitrary associations between different combinations of object stimuli (Bennett et al., 1994; Day et al., 2003; Kesner et al., 2008; O’Reilly and Rudy, 2001).