Research Interests

Our research investigates the role of brain networks that involve the frontal, medial temporal, and parietal lobes in human memory.

This work involves inter-relating cognitive hypotheses with evidence from functional neuroimaging of healthy volunteers and from examining the effects of neurological and psychiatric disorders, and normal aging, on memory abilities.

Research in the laboratory uses a number of methods, including behavioural studies, functional neuroimaging (fMRI), electrophysiology (EEG/MEG), and brain stimulation (TMS/tDCS).

View of brain showing anterior prefrontal cortex

We are especially interested in the interaction between anterior prefrontal cortex and other regions within the frontal lobes, and how these areas are involved in the strategic control of recollection (Simons, Gilbert, et al., 2005, abstract; Simons, Davis, et al., 2006, abstract; Simons, Henson, et al., 2008, abstract; Turner, Simons, et al., 2008, abstract; Gilbert, Henson, & Simons, 2010, abstract). Recent findings include the observation that structural variability in an anterior prefrontal cortex brain fold called the paracingulate sulcus might underlie individual differences in recollection that are seen in the general population (Buda, et al., 2011, abstract).

A further major focus is whether these and other brain regions play a necessary role in memory. For example, we have recently been investigating the human parietal lobe, which is commonly activated in neuroimaging experiments but which lesion studies indicate may not be critical for accurate recollection to occur (Simons, Peers, et al., 2008, abstract; Simons, Peers, et al., 2010, abstract). We recently contributed to a Special Section of the journal Neuropsychologia on this interesting topic (see editorial by Simons & Mayes, 2008, abstract; and empirical paper by Ally, Simons, et al., 2008, abstract). Our hypothesis to explain this conundrum is that the parietal lobe plays a role in the subjective experience of recollection, contributing to the richness and vividness of memories, and thus one's confidence in what one is remembering (Simons, Peers, et al., 2010, abstract).

Interactions between prefrontal cortex and medial temporal lobe in recollection

Regions of prefrontal cortex work together to engage and inhibit appropriate control processes for encoding and retrieving information that is stored in more posterior regions of the brain such as the medial temporal lobe.

A few years ago, we reviewed evidence which suggests that interactions between these prefrontal and medial temporal regions in particular are critical for successful remembering (Simons & Spiers, 2003, abstract). During encoding, these regions interact to provide discrete and elaborated representations that are amenable to long-term storage. At retrieval, interactions serve to specify retrieval cues, search the long-term store, and reactivate and monitor stored information.

Very little is known about the role that anterior prefrontal cortex plays in these processes. Our work suggests the region plays a general role in mediating attentional engagement between internally-generated and externally-derived information (Burgess, Simons, et al., 2005, abstract; Simons, 2008; Gilbert, Henson, & Simons, 2010, abstract).

Anterior prefrontal cortex and medial temporal lobe are central to recollection

During retrieval of information from memory, anterior prefrontal cortex plays a key role along with medial temporal lobe regions in recollecting the context in which previous events were experienced. Thus, the general role suggested for anterior prefrontal cortex in attentional processing may be expressed during a memory task as the differential allocation of attention between internally represented mnemonic information and the currently perceived stimuli that provoke its retrieval (Simons, Owen, et al., 2005, abstract; Simons, Gilbert, et al., 2005, abstract; Simons, Davis, et al., 2006, abstract).

Regions involved in discriminating perceived from imagined stimuli overlap with areas dysfunctional in schizophrenia

Lateral and medial aspects of anterior prefrontal cortex appear to be involved at different stages of the memory retrieval process. The lateral region supports processes engaged during the early retrieval specification stages of recollection, with the medial region contributing to later stages such as the post-retrieval monitoring of recovered information (Simons, Gilbert, et al., 2005, abstract; Simons, Henson, et al., 2008, abstract).

We have also shown that anterior prefrontal cortex forms part of a network with the thalamus and cerebellum that supports memory for whether items were previously witnessed in the outside world or imagined by subjects (Simons, Davis, et al., 2006, abstract). This prefrontal-thalamic-cerebellar network is among the areas that have been shown to be dysfunctional in schizophrenia, a disorder associated with symptoms like hallucinations which may be the result of difficulty discriminating between information that was internally-generated vs. perceived from the external world.

Interestingly, reduced activation in anterior prefrontal cortex in our data correlated with the likelihood of misattributing imagined items as having been perceived, the same externalization bias error as has been observed in patients with schizophrenia (Simons, Davis, et al., 2006, abstract; Simons, Henson, et al., 2008, abstract; Turner, Simons, et al., 2008, abstract) [See press release] [See article in The Times]. It may be that such memory errors can be directly linked to structural variability in an anterior prefrontal cortex brain fold called the paracingulate sulcus (PCS), which varies considerably in size in the general population, and can be reduced in schizophrenia. Absence of the PCS in healthy volunteers was associated with reduced ability to distinguish real events from those participants had imagined or been told about (Buda, et al., 2011, abstract) [See press release] [See article in The Guardian].

We have also investigated the cognitive and neural mechanisms underlying other aspects of memory function, including memory for contextual information and distortions of remembering.

Priming-related activation along the base of the brain

For example, using event-related fMRI, we examined the effects of perceptual and semantic manipulations on visual object priming in fusiform cortex, finding that the right fusiform appeared to be principally involved in processing specific visual form representations about objects, while the left fusiform additionally plays a role in processing lexical/semantic information (Simons, Koutstaal, et al., 2003, abstract). Another series of experiments investigated the involvement of regions of the frontal lobe in graded levels of recollection. We observed that activation in the left prefrontal cortex, in particular, was modulated by the degree of success in retrieving information from episodic memory (Maril, Simons, et al., 2003, abstract). During semantic retrieval, however, different patterns of activation characterised tip-of-the-tongue and feeling-of-knowing intermediate retrieval states (Maril, Simons, et al., 2005, abstract).

Activation in left prefrontal cortex mirrors graded levels of recollection

The involvement of prefrontal cortex in aspects of controlled retrieval was further investigated by manipulating perceptual influences on judgments of frequency of occurrence. The results suggested that left and right prefrontal cortex may contribute qualitatively different functions to episodic retrieval (Dobbins, Simons, & Schacter, 2004, abstract) [See commentary]. Other studies examined the links between diminished attentional resources and failures of source memory in older adults, indicating that memory for the specific source of information, and also for partial categorical details about the source, are affected by aging (Simons, Dodson, et al., 2004, abstract). We have recently followed up this experiment by examining the patterns of fMRI activation in prefrontal regions associated with partial and specific source recollection (Dodson, Simons, & Schacter, 2003).

A further aspect of our research has used neuropsychological and functional imaging methods to investigate the cognitive and neural organisation of different forms of long-term memory: episodic memory (personally experienced events), and semantic memory (general knowledge about objects, people, facts, concepts and word meanings).

Coronal MRI scan highlighting the hippocampal complex

To understand the mechanisms involved in these different forms of memory, we studied the performance on various experimental tasks of patients presumed to have progressive brain disorders such as Alzheimer's disease (which initially affects the hippocampal complex, associated with the acquisition and, perhaps, temporary storage of episodic memories) and semantic dementia (which primarily affects the inferior temporal lobe, thought to be the permanent store of our semantic and autobiographical memories).

In their early stages, these disorders appear, in some ways, to be mirror images of each other. While patients with early Alzheimer's disease show amnesia for day-to-day events as their initial presenting sign, our work demonstrated that patients with semantic dementia typically show highly accurate perceptually-based learning of new episodic memories (reviewed in Simons & Graham, 2000, abstract).

Multiple input hypothesis of episodic memory

We suggested that these results support a model of human long-term memory in which episodic memory draws upon multiple inputs from perceptual and semantic systems. This multiple input hypothesis can account for the preserved recognition memory in semantic dementia for pictures of objects and animals (Graham, Simons, et al., 2000, abstract; Simons, Graham, & Hodges, 2002, abstract), as well as for photographs of faces (Simons, Graham, et al., 2001-a, abstract). Additionally, we confirmed that this ability was not confined to recognition memory, but was also evidenced on tests of recollection such as source and associative memory (Simons, Verfaellie, et al., 2002, abstract). Recently, we examined false recognition in semantic dementia, finding a diminished tendency relative to controls to remember novel pictures of semantically-related objects as having been previously encountered. This reduced false recognition was specific to semantic representations, as the patients were not significantly impaired on a similar task involving categories of abstract objects (Simons, Lee, et al., 2005, abstract).

The neural bases of these effects were investigated in a series of studies that used structural and functional imaging techniques. Recognition memory in semantic dementia was shown to be affected by atrophy in the medial temporal lobe (especially the perirhinal cortex region), bilaterally for object stimuli (Simons, Graham, & Hodges, 2002, abstract) and mainly in the right hemisphere for faces (Simons, Graham, et al., 2001-a, abstract). The importance of frontal and temporal lobe areas in mediating recognition memory processes was emphasised in a functional neuroimaging study which demonstrated that memory for different stimulus types engages different areas of the brain (Simons, Graham, et al., 2001-b, abstract). Evidence suggested that prefrontal cortex was responsible for strategic aspects of memory, a notion given support by neuropsychological experiments indicating that disruption to frontal lobe function in semantic dementia, even more than atrophy to the medial temporal lobe, affected recollection-based memory (Simons, Verfaellie, et al., 2002, abstract). Such frontal lobe dysfunction did not, however, appear to correlate with ability to recall autobiographical memories from the past (Nestor, Graham, et al., 2002, abstract).

See full list of publications.