Lane et al. have done a valuable service analyzing therapeutic interventions that trade on the conceptual re-categorization of trauma experiences. Their work sheds new light on how memory-updating processes contribute to the mechanisms underlying positive therapeutic change.

Although we applaud their efforts, there are aspects of their model – in particular their conceptualization of memory – we feel merit attention. We focus on an issue we feel to be of critical importance – that is, to what do the authors refer by the terms episodic and semantic memory? Does their usage cut nature at its Platonic joints?

Because their model pivots on these constructs, conceptual precision is a matter of focal concern. We limit our critique to discussion of the conceptual criteria the authors adopt to differentiate the contributions of episodic and semantic memory to the therapeutic process. A salutary consequence of this focus is that, in the process, definitional concerns also are given a voice.

In 1985, Tulving found it necessary to downplay his original three-pronged criterion of memory-types (whereas episodic memory contains spatial and self-referential content, but semantic memory largely is devoid of such contextual content; Tulving Reference Tulving, Tulving and Donaldson1972) in favor of a partitioning predicated on the manner in which content is made available to consciousness (i.e., his autonoetic/noetic awareness distinction; e.g., Tulving Reference Tulving1985; Reference Tulving, Terrace and Metcalfe2005). One problem with the original criteria was that since the early 1980s it repeatedly had been demonstrated that self-referential, spatial, and temporal information can characterize both episodic and semantic memory (for recent review, see Klein Reference Klein2013). Accordingly, a dichotomy between semantic and episodic memory admits to considerable ambiguity when analysis of mental content serves as the basis for categorization. For example, I can know that I drove past a bookstore in Flagstaff, Arizona, on my way to the Grand Canyon in 1989 (i.e., knowledge contextualized with respect to time, space, and self) without being able to episodically recollect (i.e., re-live) the act of having done so.

Although Lane et al. mention Tulving's (Reference Tulving1985) experiential revision, they frequently revert to the idea that episodic and semantic memories are empirically separable in virtue of the content made available to awareness. This simply will not do. It is the manner in which autonoetic and noetic awareness are conjoined with content during the act of retrieval, not the content per se, which determines how a memory experience is categorized (Klein Reference Klein2013; Markowitsch & Staniloiu Reference Markowitsch and Staniloiu2011b).

Based on these concerns, it becomes apparent why a number of the authors' assertions – for example, that episodic memory (or as we currently favor to term it: episodic-autobiographical memory; Markowitsch & Staniloiu Reference Markowitsch and Staniloiu2011b; Reference Markowitsch and Staniloiu2012) entails recollection of “events” – appear unwarranted. For example, although “single events” can be the target content of recollection, mental states that do not entail memory of events can as well. Thus, I can episodically recollect that a word appeared on a list (as opposed to simply knowing that it appeared, or feeling it to be familiar). This hardly does justice to the meaning of “event.” Problems such as this stem from a failure to distinguish the content of an occurrent mental state from the manner in which that state is given to awareness (e.g., Klein Reference Klein2013, Reference Klein2014; Markowitsch & Staniloiu Reference Markowitsch and Staniloiu2011b): It is the autonoetic aspect of retrieval, not simply the content of retrieval, that makes a mental state an episodic (-autobiographical) experience (see Fig. 1).

Figure 1.

Sketch of the division into semantic and episodic-autobiographical memory. The figure includes ideas from Klein and Nichols (Reference Klein and Nichols2012), Markowitsch and Staniloiu (Reference Markowitsch and Staniloiu2012), and Picard et al. (Reference Picard, Mayor-Dubois, Maeder, Kalenzaga, Abram, Duval, Eustache, Roulet-Perez and Piolino2013).

Consider, for example, individuals who suffer profound retrograde episodic-autobiographical amnesia, such as patient A. Z. (Markowitsch & Staniloiu Reference Markowitsch and Staniloiu2013). Despite of his inability to episodically recollect any personal experiences, he was able to re-learn specific temporal and spatial details of his personal past and also knew details about celebrities. However, he experienced this content as semantic facts rather than episodic-autobiographical recollections. Similarly, Klein and Nichols (Reference Klein and Nichols2012) presented patient R. B., who temporarily lost his ability to conjoin autoneotic awareness with occurrent mental content. He could remember richly contextual details of his past experiences, but he did not take those details to be personal memory (he stated his “memory” experiences felt unowned – they seemed like facts he had been told by others, lacking the warmth and intimacy associated with episodic-autobiographical recollection; (e.g., James Reference James1890). However, when his autonoetic abilities returned, these same details were now experienced as re-living personal memories.

To generalize, in both organic and dissociative amnesia there is overwhelming evidence for a differentiation between largely preserved semantic memory (see Fig. 1) and largely impaired episodic-autobiographical memory (Markowitsch & Staniloiu Reference Markowitsch and Staniloiu2012; Staniloiu & Markowitsch Reference Staniloiu and Markowitsch2014). This general observation furthermore strongly supports Tulving's (Reference Tulving1985) distinction between these two memory systems and speaks against the use of the term “‘declarative memory’” as an umbrella of the two (cf. also Tulving & Markowitsch Reference Tulving and Markowitsch1998).

Of course, Tulving's autonoetic/noetic criterion presents serious – though not intractable (e.g., the remember/know paradigm) – difficulties for empiricism, because it trades on the subjectivity of the remembering agent. However, current methodological limitations should not sanction avoidance of core properties of the construct(s) under scrutiny (e.g., Klein Reference Klein2014). If current methods are inadequate, the appropriate scientific move is to adapt method, not to excuse foundational aspects of a construct from empirical analysis.

In conclusion, the authors' model adds to existing work on the memorial underpinnings of traumatic disorders in important ways. Although other researchers have focused on the etiology and therapeutic resolution of the traumatic effects of disturbing memories (e.g., Brewin et al. Reference Brewin, Gregory, Lipton and Burgess2010; Ehlers & Clark Reference Ehlers and Clark2000; Rubin et al. Reference Rubin, Berntsen and Bohni2008), the reconsolidation hypothesis presents a more sophisticated treatment of the memorial contributions to both the trauma experience and the recovery process.

However, because the model is grounded by the notions of episodic(-autobiographical) and semantic memory, it is incumbent on the authors' to provide a conceptually coherent and empirically warranted treatment of these focal constructs. Absent such a clarification, the present model, though perhaps therapeutically efficacious, ultimately will be found limited in its generalizability by the failure to situate its foundational constructs in a solid theoretical framework.