The first paragraph of Mahr & Csibra's (M&C's) target article claims that Tulving's semantic versus episodic distinction “has become virtually universally accepted.” From its earliest introduction to the present, this distinction has been controversial and rejected by many researchers (e.g., Anderson & Ross Reference Anderson and Ross1980; McKoon & Ratcliff Reference McKoon and Ratcliff1979) and most of the commentaries in Tulving's Behavioral and Brain Science (BBS) article (Tulving Reference Tulving1984). A more recent BBS article (Lane et al. Reference Lane, Ryan, Nadel and Greenberg2015, p. 48) notes that “the dividing line between semantic and episodic memory is getting harder to see.” For some of us, it is impossible (Rubin Reference Rubin2006). This distinction is a weak scientific basis for investigating why we remember; moreover, it is not needed.

In the target article, the main work of the semantic-episodic distinction is to justify the concept of autonoetic consciousness. Autonoetic consciousness was introduced after the basic episodic-semantic distinction (Tulving Reference Tulving1985) and can stand independent of the rest of the claims of that distinction. Similar concepts, such as recollection, have long been included in many theories of memory. Moreover, autonoetic consciousness need not depend on episodic processes. In older adults, compared to younger adults, the subjective sense of vividness of episodic memories that is a part of autonoetic consciousness can depend more on neural systems related to sematic category representations (Johnson et al. Reference Johnson, Kuhl, Mitchell, Ankudowich and Durbin2015).

The first paragraph of the target article also claims that memory research has been “preoccupied with studying the capabilities of human memory rather than aiming to illuminate its function.” However, the functions considered in the target article are limited to a small subset of intellectual traditions. M&C do not cite serious, influential researchers as alternative approaches. These include Berntsen, Bluck, Boals, McAdams, Habermas, and Tomasello (e.g., Bluck et al. Reference Bluck, Habermas and Rubin2005; Boals & Klein Reference Boals and Klein2005; Harris et al. Reference Harris, Rasmussen and Berntsen2014; McAdams Reference McAdams2006; Reference McAdams2013; Rasmussen & Berntsen Reference Rasmussen and Berntsen2009; Tomasello Reference Tomasello1999; Tomasello et al. Reference Tomasello, Carpenter, Call, Behne and Moll2005). Studies of the theory of mind that address the functions of memory, including the communicative function, are also not mentioned. The goals of the target article would be better served by a broad examination of its key communicative function concept.

M&C wrongly claim in their abstract that, in considering “episodic memory functions,” researchers have “focused on explaining its constructive character through its role in future-oriented mental time travel,” a recent and relatively minor influence on the development of theories of constructive memory. The long history of theories of the constructive nature of memory is minimized (e.g., Bartlett Reference Bartlett1932; Barsalou Reference Barsalou and Neisser1988; Brewer Reference Brewer and Rubin1986; Neisser Reference Neisser1967; Reference Neisser1976; Neisser & Fivush Reference Neisser and Fivush1994; for reviews, see Rubin Reference Rubin2006; Reference Rubin2014). The integration of these and similar theories could increase our understanding of the communicative function.

There is also a great deal known about the neural basis of memory and especially autobiographical memory that is relevant to the target article. These include papers that distinguish between laboratory-episodic and autobiographical memory studies that vary on communicative function in ways that would inform the key issues of the target article (e.g., Cabeza et al. Reference Cabeza, Prince, Daselaar, Greenberg, Budde, Dolcos, LaBar and Rubin2004; Cabeza & St. Jacques Reference Cabeza and Jacques2007; McDermott et al. Reference McDermott, Szpunar and Christ2009; St. Jacques Reference St. Jacques, Berntsen and Rubin2012). In summary, the target article omits major theoretical and empirical work on episodic memory, memory functions, the constructive nature of memory, and the neural basis of memory. Moreover, it does not fairly represent many areas it includes. I focus on one area related to my own work.

In section 1.1.3, on event memory and episodic memory, M&C review a collection of theories that share little beyond the use of the word “event.” They note that different formulations of event memory are deficient in one, but not all, of the following: location in subjective time, conscious awareness, self-reference, source information, and narrative structure. M&C put them all into an idea blender and make a homogenized, conceptual smoothie that retains nothing that could be of any theoretical use. There is a failure to note that one theory of event memory, which the target article cites (Rubin & Umanath Reference Rubin and Umanath2015), aims to replace the role played by episodic memory with event memories, defined as memories including scenes, as the key scientifically based distinction. It keeps episodic memory as a special case, which has a whole host of other properties (Tulving Reference Tulving1984, p. 224, Table 1; Rubin & Umanath Reference Rubin and Umanath2015, p. 3, Table 2), not all of which are related to the communicative function. Unlike the properties of event memory listed as contrasting to episodic memory in Table 1 of the target article, it claims event memories (i.e., memories with scenes) can be epistemically generative and autonoetic.

Amnesia and the loss of the ability to imagine a generic scene, such as a beach on a sunny day, co-occur when the neural damage is to the hippocampi (e.g., Hassabis et al. Reference Hassabis, Kumaran, Vann and Maguire2007b; Maguire & Mullally Reference Maguire and Mullally2013; Tulving Reference Tulving1985) or earlier in the visual ventral stream (Greenberg et al. Reference Greenberg, Eacott, Brechin and Rubin2005; Rubin & Greenberg Reference Rubin and Greenberg1998). These neuropsychological studies are supported by functional magnetic resonance imaging (fMRI) activation studies (e.g., Epstein & Kanwisher Reference Epstein and Kanwisher1998; Mullally & Maguire Reference Mullally and Maguire2013). Therefore, scene construction has a much clearer neural basis than episodic memory and a clear and, until recently, unexpected role in amnesia. The mental construction of any scene must be done from a specific location and time, which introduces a self located in space and time. In terms of the theory of event memory, this self is a necessary, but need not be a sufficient, condition for a sense of reliving and for belief that depends on a recollected memory of a real or imagined event. Thus, the construction of real and imagined scenes is central to communicating autobiographical memories that are perceived as relived and accurate. Visual scenes may play a role in communicative function that is at least as important as aspects of memory the target article considers.