The ventral lateral parietal cortex (VLPC) is one of the most active regions during successful episodic retrieval (reviewed by Levy Reference Levy2012; Rugg & King Reference Rugg and King2018; Sestieri et al. Reference Sestieri, Shulman and Corbetta2017; Shimamura Reference Shimamura2011; Vilberg & Rugg Reference Vilberg and Rugg2008; Wagner et al. Reference Wagner, Shannon, Kahn and Buckner2005). Nevertheless, patients with VLPC lesions can often successfully retrieve episodic memories, and are not usually considered to be amnesic. This alleged discrepancy has led to growing interest in VLPC activation in the context of episodic memory.
In considering the contribution of the VLPC to episodic retrieval, the integrative memory model proposed in the target article by Bastin et al. builds on the attention-to-memory account (AtoM; Cabeza et al. Reference Cabeza, Ciaramelli, Olson and Moscovitch2008). AtoM associates VLPC activation during retrieval with bottom-up capture of attention by relevant memory cues and/or recovered memories. Accordingly, the integrative memory model suggests that the VLPC interacts with a connectivity hub (centred in the posterior cingulate) and a frontal attribution system, to support orientation of ecphory-related attention. Nevertheless, growing evidence suggests that the role of the VLPC in retrieval goes beyond the ancillary attentional function ascribed by the integrative memory model. More specifically, building on two separate lines of evidence – the first showing VLPC involvement in content representation and the second in the evaluation of the memory trace – I suggest that the VLPC should be considered an essential part of the connectivity hub that links together the core systems and the attribution system.
Recent evidence suggest that VLPC activation reflects retrieved information, either by holding an actual representation, or by linking distributed memory traces. Particularly compelling is evidence from studies that employed multi-voxel pattern classification to decode the content of the retrieved information in the VLPC. For example, Kuhl and Chun (Reference Kuhl and Chun2014) employed a task in which words were paired with pictures (faces/scenes) during an initial study phase, and subsequently used as cues in a cued-recall test, followed by a recognition test for pictures alone. Activity patterns elicited by word cues during recall were compared with activity patterns elicited by pictures during recognition. Strikingly, in the VLPC, patterns elicited by words were more similar to the specific pictures with which they were studied than with “unassociated” pictures from the same category, strongly suggesting that the VLPC holds event-specific representations. Arguably, however, these retrieved representations are not passively held in the VLPC. Rather, their content is further evaluated and transformed into signals that can be used by the attribution system.
An influential account (Yazar et al. Reference Yazar, Bergström and Simons2012) posits that the VLPC is involved in subjective aspects of recollection. This account derives from several studies showing that, while accuracy of recognition judgements is typically unaffected by VLPC lesions, patients consistently express lower confidence in their judgements, and are less likely to report that recognised items engendered a subjective experience of recollection (e.g., Hower et al. Reference Hower, Wixted, Berryhill and Olson2014; Simons et al. Reference Simons, Peers, Mazuz, Berryhill and Olson2010). Thus, while “objective” memory performance, as expressed in response accuracy, exhibits no obvious decline in patients with VLPC lesions, “subjective” memory – that is, the personal experience of one's own episodic memory – is impaired. In support of this view, fMRI (functional magnetic resonance imaging) studies have shown that the relative number of “remember” (vs. “know”) responses, high-confidence responses, and measures of richness, vividness, and specificity of retrieved episodic events, all correlate with VLPC activation (Qin et al. Reference Qin, van Marle, Hermans and Fernandez2011; Richter et al. Reference Richter, Cooper, Bays and Simons2016; Tibon et al. Reference Tibon, Fuhrmann, Levy, Simons and Henson2019; Yazar et al. Reference Yazar, Bergström and Simons2014). Taken together, these findings suggest that the VLPC is involved in the subjective evaluation of memory traces, and affords conscious access to the quality of the memory signal that serves as the basis for such judgements (e.g., Rugg & King Reference Rugg and King2018).
The integrative memory model suggests that the connectivity between core systems and the attribution system relies mostly on the posterior cingulate cortex. Of particular interest is the ventral posterior cingulate cortex (vPCC), which connects notably with the VLPC, and is assumed to support pattern completion of a whole memory trace by allowing the reactivation of the self-referential properties of personally experienced events. Arguably, these recovered traces are then processed by the VLPC, which computes the subjective evaluation of the trace (e.g., vivid memories will be evaluated as highly confident; memories that contain many contextual details will be evaluated as recollective). The memory trace, coupled with its evaluation, is then transferred to the attribution system, which translates the signal according to task demands and particular contexts (e.g., the memory will be endorsed as “Remembered”). Altogether, the interactions between these components contribute to the quality of recollection and the subjective experience of remembering.
In contrast to the integrative memory model, the current view suggests that the frontal attribution system relies mostly on VLPC output, which represents the evaluated memory trace, rather than on vPCC output. Moreover, the interactions between the vPCC and the VLPC are (generally) hierarchical: the vPCC generates self-referenced memory traces, which are subsequently evaluated by the VLPC.
Interestingly, a recent study confirms this suggested role of the VLPC in the intersection of the core and attribution systems: Following initial recognition of studied words, participants made a remember/know judgement, and then recalled the colour and the spatial position in which the word was studied. Importantly, on trials where both features were retrieved (but not on trials where one or neither source feature was retrieved) healthy controls were more likely to make Remember relative to Know judgements, whereas parietal patients could not do so. In their interpretation, Ciaramelli et al. (Reference Ciaramelli, Faggi, Scarpazza, Mattioli, Spaniol, Ghetti and Moscovitch2017) argue that unlike patients, controls were able to use the richness of the experience accompanying the reinstatement of multiple features as an important basis for endorsing an item as “Remembered.”
To conclude, the integrative memory model integrates a large corpus of findings and theories, and provides a framework that affords better understanding of memory deficits. It will, nevertheless, benefit from further specification of the processes that occur at the intersection of the core and attribution systems; some of which are supported by VLPC.
The ventral lateral parietal cortex (VLPC) is one of the most active regions during successful episodic retrieval (reviewed by Levy Reference Levy2012; Rugg & King Reference Rugg and King2018; Sestieri et al. Reference Sestieri, Shulman and Corbetta2017; Shimamura Reference Shimamura2011; Vilberg & Rugg Reference Vilberg and Rugg2008; Wagner et al. Reference Wagner, Shannon, Kahn and Buckner2005). Nevertheless, patients with VLPC lesions can often successfully retrieve episodic memories, and are not usually considered to be amnesic. This alleged discrepancy has led to growing interest in VLPC activation in the context of episodic memory.
In considering the contribution of the VLPC to episodic retrieval, the integrative memory model proposed in the target article by Bastin et al. builds on the attention-to-memory account (AtoM; Cabeza et al. Reference Cabeza, Ciaramelli, Olson and Moscovitch2008). AtoM associates VLPC activation during retrieval with bottom-up capture of attention by relevant memory cues and/or recovered memories. Accordingly, the integrative memory model suggests that the VLPC interacts with a connectivity hub (centred in the posterior cingulate) and a frontal attribution system, to support orientation of ecphory-related attention. Nevertheless, growing evidence suggests that the role of the VLPC in retrieval goes beyond the ancillary attentional function ascribed by the integrative memory model. More specifically, building on two separate lines of evidence – the first showing VLPC involvement in content representation and the second in the evaluation of the memory trace – I suggest that the VLPC should be considered an essential part of the connectivity hub that links together the core systems and the attribution system.
Recent evidence suggest that VLPC activation reflects retrieved information, either by holding an actual representation, or by linking distributed memory traces. Particularly compelling is evidence from studies that employed multi-voxel pattern classification to decode the content of the retrieved information in the VLPC. For example, Kuhl and Chun (Reference Kuhl and Chun2014) employed a task in which words were paired with pictures (faces/scenes) during an initial study phase, and subsequently used as cues in a cued-recall test, followed by a recognition test for pictures alone. Activity patterns elicited by word cues during recall were compared with activity patterns elicited by pictures during recognition. Strikingly, in the VLPC, patterns elicited by words were more similar to the specific pictures with which they were studied than with “unassociated” pictures from the same category, strongly suggesting that the VLPC holds event-specific representations. Arguably, however, these retrieved representations are not passively held in the VLPC. Rather, their content is further evaluated and transformed into signals that can be used by the attribution system.
An influential account (Yazar et al. Reference Yazar, Bergström and Simons2012) posits that the VLPC is involved in subjective aspects of recollection. This account derives from several studies showing that, while accuracy of recognition judgements is typically unaffected by VLPC lesions, patients consistently express lower confidence in their judgements, and are less likely to report that recognised items engendered a subjective experience of recollection (e.g., Hower et al. Reference Hower, Wixted, Berryhill and Olson2014; Simons et al. Reference Simons, Peers, Mazuz, Berryhill and Olson2010). Thus, while “objective” memory performance, as expressed in response accuracy, exhibits no obvious decline in patients with VLPC lesions, “subjective” memory – that is, the personal experience of one's own episodic memory – is impaired. In support of this view, fMRI (functional magnetic resonance imaging) studies have shown that the relative number of “remember” (vs. “know”) responses, high-confidence responses, and measures of richness, vividness, and specificity of retrieved episodic events, all correlate with VLPC activation (Qin et al. Reference Qin, van Marle, Hermans and Fernandez2011; Richter et al. Reference Richter, Cooper, Bays and Simons2016; Tibon et al. Reference Tibon, Fuhrmann, Levy, Simons and Henson2019; Yazar et al. Reference Yazar, Bergström and Simons2014). Taken together, these findings suggest that the VLPC is involved in the subjective evaluation of memory traces, and affords conscious access to the quality of the memory signal that serves as the basis for such judgements (e.g., Rugg & King Reference Rugg and King2018).
The integrative memory model suggests that the connectivity between core systems and the attribution system relies mostly on the posterior cingulate cortex. Of particular interest is the ventral posterior cingulate cortex (vPCC), which connects notably with the VLPC, and is assumed to support pattern completion of a whole memory trace by allowing the reactivation of the self-referential properties of personally experienced events. Arguably, these recovered traces are then processed by the VLPC, which computes the subjective evaluation of the trace (e.g., vivid memories will be evaluated as highly confident; memories that contain many contextual details will be evaluated as recollective). The memory trace, coupled with its evaluation, is then transferred to the attribution system, which translates the signal according to task demands and particular contexts (e.g., the memory will be endorsed as “Remembered”). Altogether, the interactions between these components contribute to the quality of recollection and the subjective experience of remembering.
In contrast to the integrative memory model, the current view suggests that the frontal attribution system relies mostly on VLPC output, which represents the evaluated memory trace, rather than on vPCC output. Moreover, the interactions between the vPCC and the VLPC are (generally) hierarchical: the vPCC generates self-referenced memory traces, which are subsequently evaluated by the VLPC.
Interestingly, a recent study confirms this suggested role of the VLPC in the intersection of the core and attribution systems: Following initial recognition of studied words, participants made a remember/know judgement, and then recalled the colour and the spatial position in which the word was studied. Importantly, on trials where both features were retrieved (but not on trials where one or neither source feature was retrieved) healthy controls were more likely to make Remember relative to Know judgements, whereas parietal patients could not do so. In their interpretation, Ciaramelli et al. (Reference Ciaramelli, Faggi, Scarpazza, Mattioli, Spaniol, Ghetti and Moscovitch2017) argue that unlike patients, controls were able to use the richness of the experience accompanying the reinstatement of multiple features as an important basis for endorsing an item as “Remembered.”
To conclude, the integrative memory model integrates a large corpus of findings and theories, and provides a framework that affords better understanding of memory deficits. It will, nevertheless, benefit from further specification of the processes that occur at the intersection of the core and attribution systems; some of which are supported by VLPC.