Von Hippel and Trivers (VH&T) drive home a point psychologists often miss – a functional explanation cannot begin and end inside a person's head – people do not strive to “feel good” for its own sake, they feel good when they act in ways that, on average, increased their ancestors' chances of survival and reproduction. VH&T's target article underscores the theoretical functions of interdisciplinary work – broadening the significance of a generation of experimental studies (previously interpreted as a random array of apparently senseless information processing biases) while simultaneously grounding the fuzzy philosophical problem of self-deception in solid empirical findings. But their view raises two questions for us: First, is there really a “self” to be deceived? Second, are we really talking about “deception” or simply division of labor between mental modules?
VH&T do not go far enough in applying recent views of modularity. They focus on bivariate cognitive processes such as implicit versus explicit memory. But from an adaptationist perspective, important cognitive subdivisions cut along lines of content rather than process – different adaptive problems require qualitatively different sets of decision mechanisms. How a person's brain crunches information depends critically on whether he or she is thinking about attracting a mate, avoiding a fistfight, seeking status, making a friend, or caring for a child. Understanding those differences requires us to think about content and to think about divisions larger than two.
Thinking about the mind as composed of several motivational subselves, each dealing with different classes of problem content, has already begun to build bridges between research on social cognition and ideas in evolutionary biology, as well as generating a host of novel empirical findings (e.g., Kenrick et al. Reference Kenrick, Neuberg, Griskevicius, Becker and Schaller2010). For example, people in whom a self-protective motive is activated are more likely to remember angry faces, especially on male members of out-groups (who are otherwise homogenized in memory; Ackerman et al. Reference Ackerman, Shapiro, Neuberg, Kenrick, Becker, Griskevicius, Maner and Schaller2006) and to encode a neutral facial expression as anger (but only when it is expressed by an out-group male; Maner et al. Reference Maner, Kenrick, Becker, Robertson, Hofer, Neuberg, Delton, Butner and Schaller2005). Consistent with Trivers's classic theories about parental investment and sexual selection, males (but not females) in a mating frame of mind are more likely to interpret an attractive females' facial expression as expressing sexual interest, and mating-conscious males are likely to think more creatively and independently and to conspicuously display in other ways (Griskevicius et al. Reference Griskevicius, Cialdini and Kenrick2006; Maner et al. Reference Maner, Kenrick, Becker, Robertson, Hofer, Neuberg, Delton, Butner and Schaller2005; Sundie et al., in press).
In a brilliant article in the inaugural edition of Personality and Social Psychology Review, titled “Subselves,” Martindale (Reference Martindale and Wheeler1980) described how mental dissociations could be understood in rigorous cognitive terms. Building on cognitive concepts such as lateral inhibition and state dependent memory, Martindale described how the brain accomplishes parallel processing without attentional overload. Only a small portion of the information available to the brain can be consciously processed at any given time, requiring mechanisms for suppressing most of what is going on up there. At the level of single neurons, there is lateral inhibition; at the level of the whole functioning brain, Martindale proposed that we have different subselves – executive systems with preferential access to different memories and different action programs. Because it is simply impossible to have conscious access to all our memories, attitudes, and ongoing experiences, an implication of Martindale's analysis is that we are all, in a sense, dissociative personalities.
My colleagues and I have linked Martindale's analysis with the idea of functional modularity to propose a set of fundamental motivational systems – each of which serves functional priorities by linking different affective and motor programs to adaptively relevant environmental events. These functional motivation systems can lead to biased information processing that spans many of the “varieties of self-deception” proposed by VH&T. For example, men and women both selectively search for attractive members of the opposite sex, but later, uncommitted men and committed women misremember a greater frequency of attractive women (Maner et al. Reference Maner, Kenrick, Becker, Delton, Hofer, Wilbur and Neuberg2003). When looking at pictures of disfigured and healthy others, people selectively attend to photographs of disfigured others, but later confuse them with one another and do not remember them very well – a disjunction between attention and memory that may be functional because disfigurement (unlike an angry facial expression) is an invariant threat cue – it will still be there next time you encounter the person (Ackerman et al. Reference Ackerman, Becker, Mortensen, Sasaki, Neuberg and Kenrick2009). In addition to biases in attention and memory, people may be biased to interpret neutral expressions on goal-relevant social targets in functional ways. In one set of experiments, activating a mate search goal led men to selectively perceive sexual arousal in the neutral expressions of attractive members of the opposite sex, whereas activating a self-protection goal led men and women to selectively perceive anger in the neutral expressions of out-group males (Maner et al. Reference Maner, Kenrick, Becker, Robertson, Hofer, Neuberg, Delton, Butner and Schaller2005). Taken together, these findings reveal the importance of examining functional motivation systems and goal-relevant content, especially when considering biased information processing.
Thinking about cognitive-processing limitations and adaptive motivational systems demystifies the concept of self-deception. At the acquisition phase, we gather what is important to the currently active subself and discard what is not. At the encoding phase, we label what is important to the currently important subself and ignore what is not. When later dipping back into those memory bins, we ignore most of what is in there, and dig out what is functionally relevant to the currently active subself, to deal with the functional problem that is currently most salient. It's not deception, just selectivity.
Von Hippel and Trivers (VH&T) drive home a point psychologists often miss – a functional explanation cannot begin and end inside a person's head – people do not strive to “feel good” for its own sake, they feel good when they act in ways that, on average, increased their ancestors' chances of survival and reproduction. VH&T's target article underscores the theoretical functions of interdisciplinary work – broadening the significance of a generation of experimental studies (previously interpreted as a random array of apparently senseless information processing biases) while simultaneously grounding the fuzzy philosophical problem of self-deception in solid empirical findings. But their view raises two questions for us: First, is there really a “self” to be deceived? Second, are we really talking about “deception” or simply division of labor between mental modules?
VH&T do not go far enough in applying recent views of modularity. They focus on bivariate cognitive processes such as implicit versus explicit memory. But from an adaptationist perspective, important cognitive subdivisions cut along lines of content rather than process – different adaptive problems require qualitatively different sets of decision mechanisms. How a person's brain crunches information depends critically on whether he or she is thinking about attracting a mate, avoiding a fistfight, seeking status, making a friend, or caring for a child. Understanding those differences requires us to think about content and to think about divisions larger than two.
Thinking about the mind as composed of several motivational subselves, each dealing with different classes of problem content, has already begun to build bridges between research on social cognition and ideas in evolutionary biology, as well as generating a host of novel empirical findings (e.g., Kenrick et al. Reference Kenrick, Neuberg, Griskevicius, Becker and Schaller2010). For example, people in whom a self-protective motive is activated are more likely to remember angry faces, especially on male members of out-groups (who are otherwise homogenized in memory; Ackerman et al. Reference Ackerman, Shapiro, Neuberg, Kenrick, Becker, Griskevicius, Maner and Schaller2006) and to encode a neutral facial expression as anger (but only when it is expressed by an out-group male; Maner et al. Reference Maner, Kenrick, Becker, Robertson, Hofer, Neuberg, Delton, Butner and Schaller2005). Consistent with Trivers's classic theories about parental investment and sexual selection, males (but not females) in a mating frame of mind are more likely to interpret an attractive females' facial expression as expressing sexual interest, and mating-conscious males are likely to think more creatively and independently and to conspicuously display in other ways (Griskevicius et al. Reference Griskevicius, Cialdini and Kenrick2006; Maner et al. Reference Maner, Kenrick, Becker, Robertson, Hofer, Neuberg, Delton, Butner and Schaller2005; Sundie et al., in press).
In a brilliant article in the inaugural edition of Personality and Social Psychology Review, titled “Subselves,” Martindale (Reference Martindale and Wheeler1980) described how mental dissociations could be understood in rigorous cognitive terms. Building on cognitive concepts such as lateral inhibition and state dependent memory, Martindale described how the brain accomplishes parallel processing without attentional overload. Only a small portion of the information available to the brain can be consciously processed at any given time, requiring mechanisms for suppressing most of what is going on up there. At the level of single neurons, there is lateral inhibition; at the level of the whole functioning brain, Martindale proposed that we have different subselves – executive systems with preferential access to different memories and different action programs. Because it is simply impossible to have conscious access to all our memories, attitudes, and ongoing experiences, an implication of Martindale's analysis is that we are all, in a sense, dissociative personalities.
My colleagues and I have linked Martindale's analysis with the idea of functional modularity to propose a set of fundamental motivational systems – each of which serves functional priorities by linking different affective and motor programs to adaptively relevant environmental events. These functional motivation systems can lead to biased information processing that spans many of the “varieties of self-deception” proposed by VH&T. For example, men and women both selectively search for attractive members of the opposite sex, but later, uncommitted men and committed women misremember a greater frequency of attractive women (Maner et al. Reference Maner, Kenrick, Becker, Delton, Hofer, Wilbur and Neuberg2003). When looking at pictures of disfigured and healthy others, people selectively attend to photographs of disfigured others, but later confuse them with one another and do not remember them very well – a disjunction between attention and memory that may be functional because disfigurement (unlike an angry facial expression) is an invariant threat cue – it will still be there next time you encounter the person (Ackerman et al. Reference Ackerman, Becker, Mortensen, Sasaki, Neuberg and Kenrick2009). In addition to biases in attention and memory, people may be biased to interpret neutral expressions on goal-relevant social targets in functional ways. In one set of experiments, activating a mate search goal led men to selectively perceive sexual arousal in the neutral expressions of attractive members of the opposite sex, whereas activating a self-protection goal led men and women to selectively perceive anger in the neutral expressions of out-group males (Maner et al. Reference Maner, Kenrick, Becker, Robertson, Hofer, Neuberg, Delton, Butner and Schaller2005). Taken together, these findings reveal the importance of examining functional motivation systems and goal-relevant content, especially when considering biased information processing.
Thinking about cognitive-processing limitations and adaptive motivational systems demystifies the concept of self-deception. At the acquisition phase, we gather what is important to the currently active subself and discard what is not. At the encoding phase, we label what is important to the currently important subself and ignore what is not. When later dipping back into those memory bins, we ignore most of what is in there, and dig out what is functionally relevant to the currently active subself, to deal with the functional problem that is currently most salient. It's not deception, just selectivity.