Abstract
Scientific investigations into the functional properties of memory have recently undergone a rapid increase. These studies have revealed that processing stimuli for its survival value results in superior memory performance in children and adults. In this article, we critically evaluate this claim and conclude that survival-processing advantages in childhood and adulthood are not an indication that fitness-relevant information has adaptive priority. Instead, we argue that general memory principles (e.g., item-specific and relational processing, self-referential processing, elaboration and distinctiveness processing) better explain the adaptive function of memory. We stress the importance of these memory processes because they characterize the evolutionary adaptations of memory, are present early in life, and are developmentally invariant.
Keywords
Information processed for its importance for survival is remembered better than the same information processed in other contexts (Nairne, 2010; Otgaar, Smeets, & van Bergen, 2010). Not only does survival processing yield mnemonic benefits, but survival information itself enjoys similar advantages (Howe & Derbish, 2010; see Appendix A). These benefits are thought to arise because human (and perhaps some nonhuman animal) memory systems have evolved to prioritize survival-related information over other information because of its greater adaptive significance.
Although survival-related effects are well documented in adults, questions remain about the importance of the discovery that survival-processing effects appear early in development (Aslan & Bäuml, 2012; Otgaar, Howe, Smeets, & Garner, 2012; Otgaar & Smeets, 2010; see Appendix B). These questions are important to theories of adaptive memory 1 because many of the adaptive behaviors we see in adults, ones that have arisen through ancestral selection pressures, can and should be found in children (Bjorklund & Pelligrini, 2000; Volk & Atkinson, 2008). Indeed, for some, a particularly critical test of the adaptive nature of a specific behavior is whether it is present in younger members of the species. 2
However appealing this idea may seem, we argue that answers to ontological questions concerning the beginnings of fitness-relevant mnemonic advantages are not important to theories of adaptive memory per se. To be specific, we are referring only to the fitness-relevant functions of memory, although similar arguments might apply to fitness-irrelevant aspects of children’s memory that may also be adaptive (e.g., context-dependent memory, infantile and childhood amnesia, suggestibility). Although questions concerning age-related changes in adaptive memory are important in the context of understanding the development of memory more generally, they are perhaps of limited value when determining the origins of adaptive memory. Indeed, that our memory system has been shaped by evolutionary pressure is not in question. Neither should there be a question about the adaptive functions associated with memory from very early in life (Howe, 2011). Thus, the questions answered by developmental investigations of adaptive memory are important to a complete theory of memory development, but they do not settle issues concerning the priority of fitness-relevant information early in ontogeny.
Developmental Assumptions
Consider the assumption that biases that have evolved to retain survival-related information should already be present early in life. Although this may be true, it is equally plausible that adaptive behaviors do not emerge until later in development. Indeed, particular developmental events may have to occur prior to the appearance of an adaptive behavior (i.e., the expression of an adaptation may be experience expectant or experience dependent; Howe, 2011). Specifically, the expression of a fitness-relevant memory system may not occur until people have sufficient experience with and knowledge of survival-relevant situations. Because developmental events may differ across individuals within as well as across cultures, fitness-relevant memory may not emerge at the same time in all children. Although we may all eventually exhibit an advantage for storing and retrieving fitness-relevant information, the ontogenetic course of the emergence of this adaptive memory effect may vary considerably.
Of course, to qualify as an adaptation, a characteristic must emerge at some point during an organism’s life. However, this does not mean that adaptations need to be present early in life: Many emerge long after birth (Buss, Haselton, Shackelford, Bleske, & Wakefield, 1998; Confer et al., 2010). Examples include bipedal locomotion (approximately 1 year after birth), language (during the second year of life), and secondary sex characteristics (during puberty). Indeed, it is frequently not until this last development occurs that thoughts about sex—which are clearly relevant to adaptation, particularly reproductive behavior—emerge. The fact that thoughts about sex and the biological mechanisms necessary for reproduction do not emerge until puberty would not lead one to argue that these behaviors and thoughts are not adaptive.
To return to the memory issues at hand, what these examples illustrate is that the emergence of a behavior or characteristic, if it is adaptive, is guaranteed during the life of an organism. However, the timing of its emergence does not define whether a particular behavior or characteristic is or is not adaptive. In terms of adaptive memory and, more specifically, survival processing, it is reasonable to assume that thoughts about, experiences with, or knowledge of survival emerge only in adulthood, when our hunter-gatherer ancestors would actually have needed to act on survival-related cognitions. These cognitions might emerge even later in some of today’s (Western) cultures, given the rather extended periods of infancy and childhood that many people experience. From this perspective, the timing of the emergence of survival-processing effects in memory does not necessarily provide critical information about the adaptive nature or evolutionary origins of this form of mnemonic benefit. 3
Proximate Mechanisms and Developmental Invariance in Adaptive Memory
If these developmental survival-processing effects are not the result of the early emergence of a fitness-relevant memory system, then what are they indicative of? Although we agree that the human memory system is the result of evolutionary pressures, we argue that survival-processing effects can be explained within the context of an adaptive memory system whose proximate mechanisms are reasonably well documented (see Appendix C). When examined in the context of proximate mechanisms, many of the important features of the adaptive memory system turn out to be developmentally invariant. For example, distinctiveness processing (e.g., of features that discriminate among stimuli) plays an important role in memory from infancy (Rovee-Collier & Cuevas, 2009) onward (Howe, 2006a, 2011). Item processing (e.g., of stimulus-specific features), relational processing (e.g., of features shared among stimuli), and self-referential processing (e.g., of information relevant for the self; Ross, Anderson, & Campbell, 2011) also appear to emerge early in memory (Rovee-Collier & Cuevas, 2009; Symons & Johnson, 1997) and have developmentally invariant characteristics across childhood (Howe, 2006a, 2011).
Critically, distinctiveness, item and relational processing, and self-referential processing are mechanisms that underlie the survival-processing advantage and that contribute to the elaborate structure of traces that are best preserved in memory. Perhaps these are the adaptive tools of memory that were carved out throughout evolution and that account for adaptive memory effects and not the storage of fitness-relevant information per se. Although much survival- and fitness-relevant information is distinctive and well elaborated in memory, the mnemonic advantage of such information may simply be a by-product of these more general memory principles. In what follows, we suggest that it is these types of processing, individually or combined, that contribute to adaptive memory effects in both adults and children.
Survival-processing tasks require participants to engage in both item-specific processing (e.g., encoding information about a specific predator—for instance, that it has large eyes and attacks at night) and relational processing (e.g., encoding information about how that predator is like other predators— for instance, that it attacks from the sky and has large talons). In contrast, control-processing tasks usually require participants to engage only in one or the other type of processing. Although each type of processing enhances retention, their combination has especially powerful effects on retention. Whereas relational processing enhances retention by providing an integrated structure to organize items within a memory trace, item-specific processing enhances the discriminability (i.e., distinctiveness) of items within the trace (Burns, 2006). Together, these effects enhance the storage and retrieval of information, resulting in better memory performance in adults (Burns, 2006) and in children (Howe, 2006b). Indeed, survival tasks induce high levels of self-referential processing, and this processing enhances memory performance.
The role of self-reference in the survival-processing effect has been addressed in previous research (Nairne & Pandeirada, 2010; Nairne, Pandeirada, Gregory, & van Arsdall, 2009; Nairne, Pandeirada, & Thompson, 2008). Some of this research showed that the advantage remained among participants in a survival-processing group when they were directly compared with participants in a self-reference condition. However, the tasks in these conditions simply required participants to rate how easily a word brought to mind an important personal experience, an instruction that often fails to find a self-reference effect (Klein, in press; Klein, Loftus, & Burton, 1989). When participants are first asked to retrieve an episodic memory and then rate how easily the words bring to mind a personal experience, the survival-processing effect is eliminated. Therefore, self-referential processing is an important contributor to the survival-processing effect.
The idea that item-specific and relational processing might also underlie the survival-processing advantage was examined in a series of experiments (Burns, Hwang, & Burns, 2011). When control tasks involved only a single processing type (item specific vs. relational), survival processing produced better memory performance. However, when the control task involved both types of processing, no mnemonic advantage was found among participants who performed the survival-condition memory task. What this finding suggests is that both item-specific processing and relational processing might underlie the survival-processing advantage.
Elaboration and distinctiveness effects have been found to account for survival-processing effects in other experiments as well. If greater unusualness of the survival-task instructions leads to greater elaborative processing, then more relational processing should occur, which should lead to superior memory performance. Indeed, many of the control-condition scenarios (e.g., moving to a new home, going on a holiday) used in adaptive-memory experiments tend to be more ordinary and familiar to participants than survival scenarios, and a growing body of evidence has indicated that increased elaboration might contribute to the survival-processing effect. For example, when the richness (i.e., elaborateness) of encoding and item distinctiveness are comparable, performance differences between survival-processing and control groups disappear (Kroneisen & Erdfelder, 2011; Kroneisen, Erdfelder, & Buchner, in press).
Similarly, when the degree of elaboration is equated across rating-task scenarios (regardless of whether the resulting levels of elaboration are high or low), there is no performance advantage associated with survival processing (Howe & Derbish, in press). Specifically, when performance in the survival scenario was compared with performance in control scenarios that were unusual and ancestrally irrelevant (e.g., traveling to a foreign planet, exploring an underwater city), the typical memory advantage disappeared (Howe & Derbish, in press; Kostic, McFarlan, & Cleary, 2012). Indeed, memory performance in the standard survival scenario can be worse when compared with performance in extremely unusual scenarios (i.e., being attacked by zombies in a city or in grasslands; Soderstrom & McCabe, 2011).
Overall, these studies have shown that at least part of survival processing’s effect on memory can be attributed to elaborative processing, regardless of the ancestral relevance of the survival scenario. Indeed, the common feature among the ever-increasing number of studies that have not found such a survival-processing advantage is that they have all provided direct tests of well-known, general memory mechanisms that may have arisen because of evolutionary pressures on retention, pressures that are reflected by the mnemonic demands imposed by survival-processing tasks (Seamon et al., 2012).
Together, these studies have shown that survival processing enhances memory more than some other tasks because it recruits a powerful set of memory processes. These processes (item-specific and relational processing, self-referential processing, elaboration and distinctiveness processing) account for a whole host of memory phenomena, including the advantages associated with processing fitness-relevant information. Indeed, “it is not the evolutionary significance of survival per se that explains the survival processing effect. Rather, the degree to which survival processing invites elaborative, distinctive forms of encoding would predict the mnemonic benefit of survival processing” (Kroneisen & Erdfleder, 2011, p. 1554). Critically, the fact that these effects are developmentally invariant (Howe, 2006a, 2006b, 2011) does not allow one to conclude that survival-processing advantages in childhood imply that fitness-relevant information has adaptive priority. Rather, it shows that the proximate mechanisms (item-specific and relational processing, self-referential processing, elaboration and distinctiveness processing) that constitute the evolutionary adaptations in human memory are present early in life and are, indeed, developmentally invariant.
The synopsis that we have presented so far has mainly centered on the concept of adaptive memory as illustrated through research using the survival-processing paradigm. Of course, the evolutionary crafting of memory has also received scientific attention in research using other processes, paradigms, and perspectives (see Appendix D). For example, faces that are experienced as untrustworthy are better remembered than faces perceived as trustworthy (Rule, Slepian, & Ambady, 2012), a finding that is related to the possibility of an evolved “cheater-detection module” (Cosmides, 1989). Also, myriad studies have shown that face recognition is a highly unique ability that offers humans an obvious advantage in survival (Maguinnes & Newell, in press; but see Savine, Scullin, & Roediger, 2011). Research has also shown that the increased rates of false memories that can occur in survival-processing tasks and with survival-related materials can have positive and adaptive consequences. Indeed, Howe, Garner, Charlesworth, and Knott (2011) showed that memory illusions were able to prime solutions on complex problem-solving tasks in both children and adults. Overall, our stance is that adopting an evolutionary view of such findings has significant value, yet it is even more valuable when one considers that certain well-studied memory principles are likely to underpin these effects.
Conclusion
We have critically evaluated claims that developmental studies of memory for fitness-relevant information provide a touchstone for evolutionary models of adaptive memory. Specifically, we have argued that accounts according to which there exists a preference to store and process survival-related information early in life do not necessarily tell the whole story. In fact, there are many developmental outcomes that could be consistent with the assumption that the human memory system has been shaped by evolutionary pressures and that no one outcome is diagnostic of this fact. Indeed, many adaptations emerge long after birth (e.g., language, secondary sex characteristics), and it is equally plausible that the mnemonic advantages associated with survival processing could occur later in life, when individuals have more experience with and knowledge of fitness-relevant behaviors and situations. We have argued that many of the mnemonic processes (item-specific and relational processing, self-referential processing, elaboration and distinctiveness processing) that have been found to account for performance in survival-processing tasks also provide a compelling explanation for the developmentally invariant advantages observed for fitness-relevant information in children’s memory.
Finally, we have argued that the mnemonic advantage associated with survival processing may not lie in the adaptive significance of such processing, but in the fact that such processing recruits other well-known memory processes that enhance retention. Many of these processes occur relatively automatically and can be observed in numerous contexts regardless of age group. It is undoubtedly true that our memory system is the outcome of evolutionary pressures, ones that have sculpted today’s adaptive mnemonic processes. However, that a survival-processing advantage is a specific component of an adaptive memory system, one that is developmentally invariant, is not necessarily demanded by an evolutionary hypothesis. Indeed, developmental outcomes are agnostic on this issue. Perhaps a profitable goal of future scientific study into the functional properties of memory would be to identify the possible proximate mechanisms that play a role in adaptive memory generally, and survival-processing specifically. Although we have mentioned some of these variables, the identification of additional mechanisms operating throughout development will help us understand how our adaptive memory system changes from childhood to adulthood.
Footnotes
Appendix A: Survival-Processing and Survival-Information Effects
Appendix B: Experiments on Survival-Memory Advantages in Childhood
Otgaar and Smeets (2010) were the first to address the emergence of adaptive memory effects in childhood. Eight- and 11-year-olds were asked to rate concepts for their relevance to survival, their relevance to moving, or their pleasantness. Children evinced a recall advantage similar to that seen in adults when rating words for their utility in a survival scenario as opposed to their utility in a moving scenario or their pleasantness. A recent study by Otgaar, Howe, Smeets, and Garner (2012) even found that survival-processing effects in children could be generalized to different stimuli (i.e., words and pictures) as well as different paradigms used in memory research (i.e., intentional-memory and incidental-memory paradigms).
Aslan and Bäuml (2012) used a modified survival-processing paradigm to compare recognition performance in an incidental-memory task among children who had performed a survival-processing task and children who had performed an alternative, nonadaptive processing task. In their first experiment, three age groups were tested: 4- to 6-year-olds, 7- to 8-year-olds, and 9- to 10-year-olds. Results showed the usual improvements in memory with age, such that older children outperformed younger children. More importantly, the findings indicated that, across all age groups, items rated for survival were better recognized than items rated for pleasantness or word length.
In their second experiment with participants in the same three age groups, Aslan and Bäuml (2012) had children rate items either for their relevance to a survival scenario or for their relevance to two other scenarios that were designed to control for the use of schematic and self-referential processing. One of these nonsurvival scenarios had to do with staying overnight at a friend’s house, and the other, which also controlled for valence, had to do with being forgotten at school or kindergarten. As in the first experiment, these results showed that older children remembered more than younger children did and that children’s incidental-recognition performance, regardless of their age, was superior for items rated for their relevance to survival.
Together, the findings from these experiments paint a consistent picture of children’s incidental memory for items that have undergone survival-related processing. Indeed, these results would appear to provide ample support for the idea that the usual mnemonic advantages of processing information for its survival value appear very early (between 4 and 6 years of age) in children’s recollection repertoire and remain important throughout memory development (up to 9 to 10 years of age). Thus, it may not be unreasonable to conclude that adaptive memory is functional early in life and that our memory systems have been shaped by our ancestral past to be particularly sensitive, or “tuned in,” to fitness-relevant information, even very early in childhood, before we have much experience with or knowledge of survival-related situations.
Appendix C: Ultimate Versus Proximate Explanations of Survival Processing
To properly understand behavior (e.g., improved memory following survival processing) in evolutionary terms, one should obtain both ultimate and proximate explanations (Scott-Phillips, Dickens, & West, 2011). Critically, such explanations need not stand in opposition to each other. Rather, both levels of explanation can complement each other to provide a more complete understanding of the evolution of a certain trait or behavior. We propose that the ultimate explanation of adaptive memory can be relevant for the scientific study of memory only if one takes into account which proximate mechanisms mediate the survival-processing effect.
Appendix D: Related Adaptive Memory Processes and Paradigms
Recommended Reading
Howe, M. L. (2011). (See References). A book outlining an adaptive theory of memory and its development.
Klein, S. B. (in press). (See References). A chapter providing an excellent exegesis of the role of the self in planning and episodic future thought.
McCormack, T., & Atance, C. M. (2011). (See References). An article providing an excellent overview of the development of the planning aspects of adaptive memory.
Nairne, J. S. (2010). (See References). An article providing an excellent review of the theory and research on survival-processing effects and adaptive memory.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.
Funding
The writing of this article was supported by Grant RES-062-023-3327 from Economic and Social Research Council to M. L. H. and by Grant NWO 016.135.052 from the Netherlands Organisation for Scientific Research and a grant from the Edmond Hustinx Foundation to H. O.
