Cimpian & Salomon (C&S) make two innovative moves: First, they highlight the commonalities among seemingly distinct psychological phenomena. Although not necessarily new (e.g., see Yzerbyt et al. Reference Yzerbyt, Rocher, Schadron, Spears, Oakes, Ellemers and Haslam1997), such an endeavor is valuable because it enables connecting different domains and possibly developing a more parsimonious understanding of human functioning. Second, they argue that these commonalities all trace back to a single underlying cognitive mechanism – that is, the inherence heuristic. Doing so, they fall into the trap of the exact same heuristic that they describe: They assume that the similarities between different psychological phenomena ought to be the reflection of some inherent human cognitive functioning. Sensible as this idea may be, we argue (1) that the empirical evidence offered by the authors is insufficient to sustain their claim; (2) that, beyond identifiable methodological problems, the direction of their proposed underlying cognitive mechanism is largely speculative and that a reverse, more motivational, path is equally plausible; and (3) that the distinction and antecedence of inherence intuitions with respect to essentialist beliefs remain to be demonstrated.
Throughout their article, C&S present a series of arguments along with allegedly supportive evidence in favor of the idea that human beings show a deep and irrepressible inclination to “make sense of observed patterns in terms of the inherent features of their constituents” (sect. 3, para. 1). Interestingly, they hasten to soothe their message and acknowledge that products of the explanatory process are not always consistent with the inherence heuristic: Next to individual and developmental differences, people are not only more prone to make dispositional attributions when observing others (or members of other groups) than when explaining their own behaviors (or the behaviors of close others), but the valence of the observed behaviors would also seem to be an important moderator of this effect (e.g., Hewstone Reference Hewstone1990). According to C&S, inconsistencies emerge when individuals block or revise the inherence heuristic process, enabling more extrinsic explanations to emerge. If this were the case, one should expect that when blocking or revising is hindered, say, because participants are under cognitive load, intuitions based on the inherence heuristic should pop up again. Such empirical evidence is not offered by C&S. Also, their strong view has some trouble with findings showing that cultural and pragmatic factors moderate the correspondence bias (e.g., Leyens et al. Reference Leyens, Yzerbyt and Corneille1996).
In addition, those studies that most directly test the authors' hypotheses are not without limitations. To mention only one, the inherence heuristic scale should be thoroughly tested for its discriminant validity with respect to closely related constructs – for example, adherence to cultural norms or conservative tendencies. If, as we suspect, the scale measures norm adherence, it is likely that children would report lower absolute levels on this scale than would adults, an outcome that would be incompatible with C&S's argument that the inherence heuristic is more prevalent during childhood.
More critically, C&S provide no clear evidence regarding the causal direction of the hypothesized inherence heuristic. They propose that information about the targets is activated first, that this information is biased toward inherence, and that the activated inherent features then give rise to inherent-type explanations for the (observed) pattern. Because of the correlational nature of the data, a reverse causal path is equally plausible, one that would start with the activation of an inherence explanation, which would lead to the subsequent preferential reliance on the inherent characteristics of the pattern's constituents (see Yzerbyt et al. Reference Yzerbyt, Corneille and Estrada2001). It is well known that human beings have a strong motivation to develop a sense that the world is coherent and predictable (Fiske & Taylor Reference Fiske and Taylor2008). Because inherent accounts are especially well suited to answer humans' need for coherence and predictability, these accounts should be preferentially selected as working hypotheses about events' co-occurrence. In this model also, inherent features of the constituents are activated, but, rather than being precursors of the inherence intuition, they are conceived of as its consequences, via a hypothesis confirmation bias (Snyder Reference Snyder and Berkowitz1984). Concretely, when observing girls' affinity with pink, people initially develop the hypothesis that there must be “something” underneath that explains the color preference (i.e., the inherence intuition) before elaborating on what this “something” could be (i.e., the constituents' inherent features – girls' feminine nature).
This reverse causal path is entirely consistent with the observation that people often develop inherence intuitions despite their inability to pinpoint directly what the inherent-type explanation may be. That is, intuitions develop before people gain access to the exact features on which they later develop their account. The reverse path also explains why inherence intuitions tend to persist even if external (e.g., historical, socioconventional) explanations are otherwise accessible. As people test the inherence hypothesis, they search for information that confirms (inherent features) rather than questions (external constraints) their a priori beliefs. Moreover, although C&S draw a clear-cut line between inherence explanations and historical/socioconventional ones, we suggest that the two types of accounts are not necessarily mutually exclusive or incompatible in people's minds. People do not always choose for one explanation by dismissing the other. Instead, when assessing the credibility of their inherence intuition, perceivers are able to interpret historical constraints in hypothesis-serving ways. (For example, the girl–pink association emerged in marketing campaigns in the nineteenth century because pink is an inherently feminine color.)
Finally, the differences and similarities between an essentialist stance and the inherence heuristic remain weakly documented at a theoretical level and, indeed, hardly supported at the empirical one. C&S report evidence that their inherence scale is strongly correlated with a standard essentialism scale (Haslam et al. Reference Haslam, Rothschild and Ernst2000), even after controlling for a host of cognitive, personality, and ideological dimensions. Notwithstanding the difficulties of appraising the exact nature of the inherence scale, such findings tell us nothing about, and even tend to undermine, the general argument that inherence intuitions precede essentialist beliefs.
Cimpian & Salomon (C&S) make two innovative moves: First, they highlight the commonalities among seemingly distinct psychological phenomena. Although not necessarily new (e.g., see Yzerbyt et al. Reference Yzerbyt, Rocher, Schadron, Spears, Oakes, Ellemers and Haslam1997), such an endeavor is valuable because it enables connecting different domains and possibly developing a more parsimonious understanding of human functioning. Second, they argue that these commonalities all trace back to a single underlying cognitive mechanism – that is, the inherence heuristic. Doing so, they fall into the trap of the exact same heuristic that they describe: They assume that the similarities between different psychological phenomena ought to be the reflection of some inherent human cognitive functioning. Sensible as this idea may be, we argue (1) that the empirical evidence offered by the authors is insufficient to sustain their claim; (2) that, beyond identifiable methodological problems, the direction of their proposed underlying cognitive mechanism is largely speculative and that a reverse, more motivational, path is equally plausible; and (3) that the distinction and antecedence of inherence intuitions with respect to essentialist beliefs remain to be demonstrated.
Throughout their article, C&S present a series of arguments along with allegedly supportive evidence in favor of the idea that human beings show a deep and irrepressible inclination to “make sense of observed patterns in terms of the inherent features of their constituents” (sect. 3, para. 1). Interestingly, they hasten to soothe their message and acknowledge that products of the explanatory process are not always consistent with the inherence heuristic: Next to individual and developmental differences, people are not only more prone to make dispositional attributions when observing others (or members of other groups) than when explaining their own behaviors (or the behaviors of close others), but the valence of the observed behaviors would also seem to be an important moderator of this effect (e.g., Hewstone Reference Hewstone1990). According to C&S, inconsistencies emerge when individuals block or revise the inherence heuristic process, enabling more extrinsic explanations to emerge. If this were the case, one should expect that when blocking or revising is hindered, say, because participants are under cognitive load, intuitions based on the inherence heuristic should pop up again. Such empirical evidence is not offered by C&S. Also, their strong view has some trouble with findings showing that cultural and pragmatic factors moderate the correspondence bias (e.g., Leyens et al. Reference Leyens, Yzerbyt and Corneille1996).
In addition, those studies that most directly test the authors' hypotheses are not without limitations. To mention only one, the inherence heuristic scale should be thoroughly tested for its discriminant validity with respect to closely related constructs – for example, adherence to cultural norms or conservative tendencies. If, as we suspect, the scale measures norm adherence, it is likely that children would report lower absolute levels on this scale than would adults, an outcome that would be incompatible with C&S's argument that the inherence heuristic is more prevalent during childhood.
More critically, C&S provide no clear evidence regarding the causal direction of the hypothesized inherence heuristic. They propose that information about the targets is activated first, that this information is biased toward inherence, and that the activated inherent features then give rise to inherent-type explanations for the (observed) pattern. Because of the correlational nature of the data, a reverse causal path is equally plausible, one that would start with the activation of an inherence explanation, which would lead to the subsequent preferential reliance on the inherent characteristics of the pattern's constituents (see Yzerbyt et al. Reference Yzerbyt, Corneille and Estrada2001). It is well known that human beings have a strong motivation to develop a sense that the world is coherent and predictable (Fiske & Taylor Reference Fiske and Taylor2008). Because inherent accounts are especially well suited to answer humans' need for coherence and predictability, these accounts should be preferentially selected as working hypotheses about events' co-occurrence. In this model also, inherent features of the constituents are activated, but, rather than being precursors of the inherence intuition, they are conceived of as its consequences, via a hypothesis confirmation bias (Snyder Reference Snyder and Berkowitz1984). Concretely, when observing girls' affinity with pink, people initially develop the hypothesis that there must be “something” underneath that explains the color preference (i.e., the inherence intuition) before elaborating on what this “something” could be (i.e., the constituents' inherent features – girls' feminine nature).
This reverse causal path is entirely consistent with the observation that people often develop inherence intuitions despite their inability to pinpoint directly what the inherent-type explanation may be. That is, intuitions develop before people gain access to the exact features on which they later develop their account. The reverse path also explains why inherence intuitions tend to persist even if external (e.g., historical, socioconventional) explanations are otherwise accessible. As people test the inherence hypothesis, they search for information that confirms (inherent features) rather than questions (external constraints) their a priori beliefs. Moreover, although C&S draw a clear-cut line between inherence explanations and historical/socioconventional ones, we suggest that the two types of accounts are not necessarily mutually exclusive or incompatible in people's minds. People do not always choose for one explanation by dismissing the other. Instead, when assessing the credibility of their inherence intuition, perceivers are able to interpret historical constraints in hypothesis-serving ways. (For example, the girl–pink association emerged in marketing campaigns in the nineteenth century because pink is an inherently feminine color.)
Finally, the differences and similarities between an essentialist stance and the inherence heuristic remain weakly documented at a theoretical level and, indeed, hardly supported at the empirical one. C&S report evidence that their inherence scale is strongly correlated with a standard essentialism scale (Haslam et al. Reference Haslam, Rothschild and Ernst2000), even after controlling for a host of cognitive, personality, and ideological dimensions. Notwithstanding the difficulties of appraising the exact nature of the inherence scale, such findings tell us nothing about, and even tend to undermine, the general argument that inherence intuitions precede essentialist beliefs.