On Barzykowski and Moulin's continuum hypothesis, the feeling of familiarity involved in déjà vu experiences and the episodic recollection involved in involuntary autobiographical memories (IAMs) are due to the same underpinning neurocognitive retrieval processes. Elaborating their hypothesis with a threshold model, the authors propose that cues in the environment constantly activate autobiographical memory. As a result, some memories cross the awareness threshold due to sufficient activation and give rise to complete retrieval. But it also happens that memories, due to weak activation for instance, give rise to incomplete retrieval, or that no specific memory crosses the threshold and no content is retrieved. In these cases, the cue is predicted to feel familiar while no complete recollection, or no recollection at all, occurs.
Though such a single-mechanism model can avail itself of some recent proposals along the same lines, as Bastin et al. (Reference Bastin, Besson, Simon, Delhaye, Geurten, Willems and Salmon2019) note, the majority of the existing models suggest that there is fractionation of both the memory processes (Yonelinas, Reference Yonelinas2002) and the dedicated brain regions (Aggleton & Brown, Reference Aggleton and Brown1999) by reference to familiarity and recollection, in particular in the medial temporal lobe, with parahippocampal regions dedicated to familiarity and the hippocampal structure dedicated to recollection. Important recent theorizing of episodic memory as a distinct system specifically involving mental time travel feelings (Barry & Maguire, Reference Barry and Maguire2019, for instance) lend further support to the same suggestion. In brief, the feeling of familiarity and episodic recollection exhibit neatly distinct neurocognitive signatures. If these in-the-majority models are right, the continuum hypothesis is in a bad position, since one of its main goals is to establish that déjà vu and IAMs can be grouped as two forms of involuntary cognition unified by the same underlying retrieval process. There is therefore a discontinuity issue for the continuum hypothesis.
I think that the hypothesis can overcome this issue. But to do so, it must make room for a distinction neglected by its current formulation, and be ready for ensuing refinements.
Though the authors make no room for this, two types of déjà experience must be distinguished, arguably. Subjects can undergo the experience of reliving in the present the very same episode of experience as one they have already lived before. The author's fourth initial example describes precisely this: “It feels like you're living an experience that you've already lived through.” But subjects can also undergo déjà experiences illustrated by reports like the following: “Frequently I relate this to seeing people in the street. I will ‘Recognise them,’ rack my brain trying to remember where and expect them to greet me as we pass” (Illman, Reference Illman2012). In such experiences, subjects have the feeling that a particular perceived item like a person – versus a whole episode of experience – has been already encountered during some indeterminate, potentially multiple past episodes of experience – versus a specific episode of experience – which refers to the feeling of familiarity on a standard characterization of it. Some have judged these differences sufficiently important to call the first type of experiences déjà vécu and distinguished it from déjà vu, arguing that while in déjà vu a feeling of familiarity occurs, in déjà vécu an episodic-recollection-related feeling occurs (O'Connor, Lever, & Moulin, Reference O'Connor, Lever and Moulin2010).
On my diagnosis, the discontinuity issue originates in the authors' attempt to account for the continuity of déjà experiences with episodic IAMs by considering déjà vu instead of considering déjà vécu. This diagnosis suggests a way out.
Though déjà vécu is akin to episodic IAM, they are distinct phenomena. Episodic IAM consists of an actual recollection of a past episode of experience. By contrast, déjà vécu consists of the awareness of the ongoing episode of experience as having been lived before with no actual recollection of any previous occurrence of it. Following a recent elaboration of this point, while episodic recollection brings into play an autonoetic feeling, déjà vécu brings into play an episodic feeling of knowing (EFOK), namely the feeling that one could remember other occurrences of the ongoing episode of experience despite not carrying out this memory task at the moment (Perrin, Moulin, & Sant'Anna, Reference Perrin, Moulin and Sant'Annaforthcoming). Interestingly, a prevailing account of EFOK says that shared processes underly both this feeling and actual episodic recollection. According to Souchay, Moulin, Clarys, Taconnat, and Isingrini (Reference Souchay, Moulin, Clarys, Taconnat and Isingrini2007), EFOK results from the subpersonal and partial retrieval of contextual details, phenomenological feelings, and self-awareness similar to the ones that occur in actual episodic recollection. My proposal is, then, that déjà vécu experience is underpinned by the same retrieval process as episodic IAM, a process that would be assessed as malfunctioning when such an experience occurs.
What about déjà vu? It is worth recalling at this stage that autobiographical memory, IAM included, comprises two main forms (Conway, Reference Conway2001). Though the authors talk mainly about the episodic form, autobiographical memory also possesses a semantic form. My proposal is, then, that while EFOK-based déjà vécu experiences share their underpinning neurocognitive processes with episodic IAMs, feeling-of-familiarity-based déjà vu experiences share their underpinning neurocognitive processes with semantic IAMs. In effect, it has since long been noted that familiarity and semantic memory are closely related. As one remembers that one has already encountered in the past an item, the item typically feels familiar, and Tulving (Reference Tulving1985) even assimilates the noetic consciousness characteristic of semantic memory to familiarity.
If those proposals are correct, the continuity issue pointed out above can be put to rest and the continuum hypothesis reinstated. As I have argued, the authors' version of the hypothesis is threatened because it considers only familiarity-based déjà experiences and posits neurocognitive continuity between this type of experiences and episodic IAMs. Once the déjà vu–déjà vécu distinction and the episodic-semantic IAMs distinction are clear, one can posit continuity relationships at the right place, namely between déjà vu experiences and semantic IAMs, on the one hand, and between episodic IAMs and déjà vécu experiences, on the other hand.
You have
Access
On Barzykowski and Moulin's continuum hypothesis, the feeling of familiarity involved in déjà vu experiences and the episodic recollection involved in involuntary autobiographical memories (IAMs) are due to the same underpinning neurocognitive retrieval processes. Elaborating their hypothesis with a threshold model, the authors propose that cues in the environment constantly activate autobiographical memory. As a result, some memories cross the awareness threshold due to sufficient activation and give rise to complete retrieval. But it also happens that memories, due to weak activation for instance, give rise to incomplete retrieval, or that no specific memory crosses the threshold and no content is retrieved. In these cases, the cue is predicted to feel familiar while no complete recollection, or no recollection at all, occurs.
Though such a single-mechanism model can avail itself of some recent proposals along the same lines, as Bastin et al. (Reference Bastin, Besson, Simon, Delhaye, Geurten, Willems and Salmon2019) note, the majority of the existing models suggest that there is fractionation of both the memory processes (Yonelinas, Reference Yonelinas2002) and the dedicated brain regions (Aggleton & Brown, Reference Aggleton and Brown1999) by reference to familiarity and recollection, in particular in the medial temporal lobe, with parahippocampal regions dedicated to familiarity and the hippocampal structure dedicated to recollection. Important recent theorizing of episodic memory as a distinct system specifically involving mental time travel feelings (Barry & Maguire, Reference Barry and Maguire2019, for instance) lend further support to the same suggestion. In brief, the feeling of familiarity and episodic recollection exhibit neatly distinct neurocognitive signatures. If these in-the-majority models are right, the continuum hypothesis is in a bad position, since one of its main goals is to establish that déjà vu and IAMs can be grouped as two forms of involuntary cognition unified by the same underlying retrieval process. There is therefore a discontinuity issue for the continuum hypothesis.
I think that the hypothesis can overcome this issue. But to do so, it must make room for a distinction neglected by its current formulation, and be ready for ensuing refinements.
Though the authors make no room for this, two types of déjà experience must be distinguished, arguably. Subjects can undergo the experience of reliving in the present the very same episode of experience as one they have already lived before. The author's fourth initial example describes precisely this: “It feels like you're living an experience that you've already lived through.” But subjects can also undergo déjà experiences illustrated by reports like the following: “Frequently I relate this to seeing people in the street. I will ‘Recognise them,’ rack my brain trying to remember where and expect them to greet me as we pass” (Illman, Reference Illman2012). In such experiences, subjects have the feeling that a particular perceived item like a person – versus a whole episode of experience – has been already encountered during some indeterminate, potentially multiple past episodes of experience – versus a specific episode of experience – which refers to the feeling of familiarity on a standard characterization of it. Some have judged these differences sufficiently important to call the first type of experiences déjà vécu and distinguished it from déjà vu, arguing that while in déjà vu a feeling of familiarity occurs, in déjà vécu an episodic-recollection-related feeling occurs (O'Connor, Lever, & Moulin, Reference O'Connor, Lever and Moulin2010).
On my diagnosis, the discontinuity issue originates in the authors' attempt to account for the continuity of déjà experiences with episodic IAMs by considering déjà vu instead of considering déjà vécu. This diagnosis suggests a way out.
Though déjà vécu is akin to episodic IAM, they are distinct phenomena. Episodic IAM consists of an actual recollection of a past episode of experience. By contrast, déjà vécu consists of the awareness of the ongoing episode of experience as having been lived before with no actual recollection of any previous occurrence of it. Following a recent elaboration of this point, while episodic recollection brings into play an autonoetic feeling, déjà vécu brings into play an episodic feeling of knowing (EFOK), namely the feeling that one could remember other occurrences of the ongoing episode of experience despite not carrying out this memory task at the moment (Perrin, Moulin, & Sant'Anna, Reference Perrin, Moulin and Sant'Annaforthcoming). Interestingly, a prevailing account of EFOK says that shared processes underly both this feeling and actual episodic recollection. According to Souchay, Moulin, Clarys, Taconnat, and Isingrini (Reference Souchay, Moulin, Clarys, Taconnat and Isingrini2007), EFOK results from the subpersonal and partial retrieval of contextual details, phenomenological feelings, and self-awareness similar to the ones that occur in actual episodic recollection. My proposal is, then, that déjà vécu experience is underpinned by the same retrieval process as episodic IAM, a process that would be assessed as malfunctioning when such an experience occurs.
What about déjà vu? It is worth recalling at this stage that autobiographical memory, IAM included, comprises two main forms (Conway, Reference Conway2001). Though the authors talk mainly about the episodic form, autobiographical memory also possesses a semantic form. My proposal is, then, that while EFOK-based déjà vécu experiences share their underpinning neurocognitive processes with episodic IAMs, feeling-of-familiarity-based déjà vu experiences share their underpinning neurocognitive processes with semantic IAMs. In effect, it has since long been noted that familiarity and semantic memory are closely related. As one remembers that one has already encountered in the past an item, the item typically feels familiar, and Tulving (Reference Tulving1985) even assimilates the noetic consciousness characteristic of semantic memory to familiarity.
If those proposals are correct, the continuity issue pointed out above can be put to rest and the continuum hypothesis reinstated. As I have argued, the authors' version of the hypothesis is threatened because it considers only familiarity-based déjà experiences and posits neurocognitive continuity between this type of experiences and episodic IAMs. Once the déjà vu–déjà vécu distinction and the episodic-semantic IAMs distinction are clear, one can posit continuity relationships at the right place, namely between déjà vu experiences and semantic IAMs, on the one hand, and between episodic IAMs and déjà vécu experiences, on the other hand.
Financial support
None.
Competing interest
None.