Tuesday, September 22, 2020

Drawing on dual-process theory, we suggest that the benefits that arise from combining several quantitative individual judgments will be heightened when the judgments are based on different cognitive processes

Enhancing the Wisdom of the Crowd With Cognitive-Process Diversity: The Benefits of Aggregating Intuitive and Analytical Judgments. Steffen Keck, Wenjie Tang. Psychological Science, September 22, 2020. https://doi.org/10.1177/0956797620941840

Rolf Degen's take: https://twitter.com/DegenRolf/status/1308645051766116352

Abstract: Drawing on dual-process theory, we suggest that the benefits that arise from combining several quantitative individual judgments will be heightened when these judgments are based on different cognitive processes. We tested this hypothesis in three experimental studies in which participants provided estimates for the dates of different historical events (Study 1, N = 152), made probabilistic forecasts for the outcomes of soccer games (Study 2, N = 98), and estimated the weight of individuals on the basis of a photograph (Study 3, N = 3,695). For each of these tasks, participants were prompted to make judgments relying on an analytical process, on their intuition, or (in a control condition) on no specific instructions. Across all three studies, our results show that an aggregation of intuitive and analytical judgments provides more accurate estimates than any other aggregation procedure and that this advantage increases with the number of aggregated judgments.

Keywords: decision making, judgment, cognitive processes, wisdom of the crowd, judgment aggregation, group judgments, dual-process theory, open data

The results of three experimental studies showed that forming crowds with a high level of cognitive-process diversity—by aggregating a combination of intuitive and analytical individual judgments—improved the quality of crowd wisdom, compared with crowds formed by an aggregation of only analytical judgments, only intuitive judgments, or judgments made in a control condition without specific manipulation of judges’ cognitive processes. Moreover, we found that whereas the benefits of cognitive-process diversity generally held for both smaller and larger crowds, the magnitude of these benefits increased with crowd size and eventually approached its maximum as crowds became very large. Providing supporting evidence for the suggestion that the benefits of cognitive-process diversity are driven by higher levels of judgment-error independence, the results of Studies 1 and 2 revealed a lower average correlation in signed errors between judges employing an intuitive cognitive process and those employing an analytical cognitive process, compared with judges relying on the same cognitive process or judges in the control condition.

One particularly interesting finding of Study 3 is that analytical-intuitive crowds still outperformed purely analytical crowds even though individual analytical judgments were more accurate than individual intuitive judgments—implying that in this specific context the benefits of adding more uncorrelated judgments outweighed the detrimental effects of adding less accurate judgments. It is, however, important to note that there are likely a number of domains (e.g., tasks that require the application of formal logic) in which intuitive judgments would be much less accurate than analytical ones and hence adding highly inaccurate though less correlated judgments to a crowd is likely not beneficial (e.g., Mannes et al., 2014).

Previous research has suggested ways to improve judgment aggregation, such as by selecting better performing individuals (e.g., Budescu & Chen, 2014Mannes et al., 2014) or by refining the aggregation procedure (e.g., Jose & Winkler, 2008Palley & Soll, 2019). By contrast, our approach focused on increasing independence between individual judgment errors by manipulating the cognitive process employed by individual judges to form their judgments. It thus also complements recent work by de Oliveira and Nisbett (2018), who investigated the possibility of improving crowd wisdom by amplifying the demographic diversity of crowds and found that this approach was largely ineffective. A likely explanation for this difference in results is that we directly manipulated the cognitive process by which judgments were being made, whereas demographic differences frequently might not be associated with differences in individual cognition.

One limitation of our work is that we manipulated judgments to either be predominantly intuitive or predominantly analytical. However, in practice, judgments and decisions might frequently be based on a process in the middle of a continuum with analytical and intuitive processes at the boundaries (e.g., Hammond, 1996). Thus, an interesting direction for future research would be to compare our approach with one in which a crowd is formed by aggregating judgments that are each based on a mixture of analytical and intuitive processes. A related important limitation of our results is that we did not provide direct insights into differences in participants’ exact cognitive processes, such as the use of different judgment rules or reliance on different pieces of information (e.g., Herzog & von Helversen, 2018Hoffmann et al., 2013). Such differences might explain the higher independence between analytical and intuitive judgments observed in our studies.

A final interesting avenue for future research would be to explore whether our approach toward improving the wisdom of crowds might also help to increase the effectiveness of combining judgments that are made by the same individual (e.g., Herzog & Hertwig, 20092014Vul & Pashler, 2008).

The results of three experimental studies showed that forming crowds with a high level of cognitive-process diversity—by aggregating a combination of intuitive and analytical individual judgments—improved the quality of crowd wisdom, compared with crowds formed by an aggregation of only analytical judgments, only intuitive judgments, or judgments made in a control condition without specific manipulation of judges’ cognitive processes. Moreover, we found that whereas the benefits of cognitive-process diversity generally held for both smaller and larger crowds, the magnitude of these benefits increased with crowd size and eventually approached its maximum as crowds became very large. Providing supporting evidence for the suggestion that the benefits of cognitive-process diversity are driven by higher levels of judgment-error independence, the results of Studies 1 and 2 revealed a lower average correlation in signed errors between judges employing an intuitive cognitive process and those employing an analytical cognitive process, compared with judges relying on the same cognitive process or judges in the control condition.

One particularly interesting finding of Study 3 is that analytical-intuitive crowds still outperformed purely analytical crowds even though individual analytical judgments were more accurate than individual intuitive judgments—implying that in this specific context the benefits of adding more uncorrelated judgments outweighed the detrimental effects of adding less accurate judgments. It is, however, important to note that there are likely a number of domains (e.g., tasks that require the application of formal logic) in which intuitive judgments would be much less accurate than analytical ones and hence adding highly inaccurate though less correlated judgments to a crowd is likely not beneficial (e.g., Mannes et al., 2014).

Previous research has suggested ways to improve judgment aggregation, such as by selecting better performing individuals (e.g., Budescu & Chen, 2014Mannes et al., 2014) or by refining the aggregation procedure (e.g., Jose & Winkler, 2008Palley & Soll, 2019). By contrast, our approach focused on increasing independence between individual judgment errors by manipulating the cognitive process employed by individual judges to form their judgments. It thus also complements recent work by de Oliveira and Nisbett (2018), who investigated the possibility of improving crowd wisdom by amplifying the demographic diversity of crowds and found that this approach was largely ineffective. A likely explanation for this difference in results is that we directly manipulated the cognitive process by which judgments were being made, whereas demographic differences frequently might not be associated with differences in individual cognition.

One limitation of our work is that we manipulated judgments to either be predominantly intuitive or predominantly analytical. However, in practice, judgments and decisions might frequently be based on a process in the middle of a continuum with analytical and intuitive processes at the boundaries (e.g., Hammond, 1996). Thus, an interesting direction for future research would be to compare our approach with one in which a crowd is formed by aggregating judgments that are each based on a mixture of analytical and intuitive processes. A related important limitation of our results is that we did not provide direct insights into differences in participants’ exact cognitive processes, such as the use of different judgment rules or reliance on different pieces of information (e.g., Herzog & von Helversen, 2018Hoffmann et al., 2013). Such differences might explain the higher independence between analytical and intuitive judgments observed in our studies.

A final interesting avenue for future research would be to explore whether our approach toward improving the wisdom of crowds might also help to increase the effectiveness of combining judgments that are made by the same individual (e.g., Herzog & Hertwig, 20092014Vul & Pashler, 2008).

Just kidding: the evolutionary roots of playful teasing

Just kidding: the evolutionary roots of playful teasing. Johanna Eckert, Sasha L. Winkler and Erica A. Cartmill. Biology Letters, September 23 2020, Volume 16, Issue 9. https://doi.org/10.1098/rsbl.2020.0370

Rolf Degen's take: https://twitter.com/DegenRolf/status/1308635232736862209

Abstract: Accounts of teasing have a long history in psychological and sociological research, yet teasing itself is vastly underdeveloped as a topic of study. As a phenomenon that moves along the border between aggression and play, teasing presents an opportunity to investigate key foundations of social and mental life. Developmental studies suggest that preverbal human infants already playfully tease their parents by performing ‘the unexpected,’ apparently deliberately violating the recipient's expectations to create a shared humorous experience. Teasing behaviour may be phylogenetically old and perhaps an evolutionary precursor to joking. In this review, we present preliminary evidence suggesting that non-human primates also exhibit playful teasing. In particular, we argue that great apes display three types of playful teasing described in preverbal human infants: teasing with offer and withdrawal, provocative non-compliance and disrupting others' activities. We highlight the potential of this behaviour to provide a window into complex socio-cognitive processes such as attribution of others’ expectations and, finally, we propose directions for future research and call for systematic studies of teasing behaviour in non-human primates.


4. Cognitive implications of playful teasing

Some authors have proposed that playful teasing in human infants provides a window into their rich early ‘theory of mind’ abilities, as well as into proto-forms of humour (e.g. [18,21,2327,64]). If apes (or other animals) engage in similar forms of playful teasing, do they also have some understanding of the expectations of others? Is it possible that great apes, like human infants, deliberately play with these expectations for the sake of amusement?

The study of ‘theory of mind,’ i.e. the ability to ascribe mental states to others, has been of central interest in comparative psychology for several decades (see [6567] for reviews). There is ample evidence showing that great apes (i) ascribe intentions and goals to others (e.g. [66,6870]), (ii) are aware of attentional states of others (i.e. what they can see or hear; e.g. [7174]) and (iii) make use of this knowledge in both competitive and cooperative contexts (e.g. [75]). Crucially, recent research demonstrated that apes are also capable of ‘mind-reading’ abilities that require a simultaneous representation of two conflicting views of the world: one's own (correct) perspective and the (incorrect) perspective of another individual [76]. Hence, great apes are not only sensitive to what other individuals intend to do and what they know, but they also have some understanding of others' beliefs, even when these beliefs conflict with reality (also see [7779] for similar findings on false belief attribution in young children).

Playful teasing events, such as the offer-withdrawal, presumably involve rich inferences on both the side of the teaser and the side of the recipient. A typical offer-withdrawal event starts with the teaser making an ‘offer’ gesture, inviting the recipient to reach for an extended object or limb. All species of great apes produce offer gestures, e.g. in the context of food sharing [80] or grooming [81]. Hence, it is reasonable to assume that both parties are aware of the typical use of this gesture to draw attention to a body part of the signaller or to transfer something to the recipient. Also, there is evidence that apes produce this gesture type intentionally to pursue a particular goal [54,82]. Gestures are typically deemed to be intentional if they are (i) motorically ineffective, (ii) directed towards another individual, (iii) goal-directed and (iv) demonstrate flexibility in their usage. Goal-directedness is often shown through the use of response waiting or through persistent attempts to communicate. Teasing events may take different forms than gestures, but if they are fundamentally communicative in nature and are aimed at eliciting a particular response from the target, they will likely demonstrate the same markers of intentionality as seen in ape gesturing.

These markers seemed to be present in the videos of orangutan offer-and-withdrawal events (collected for [55]). The teaser usually seemed to await a particular response from the recipient after offering the object or limb (anticipating a reaching-out-to-take action). If this response was not given, the teaser slightly modified or intensified the offer gesture. In one case, an orangutan offered another a stick by holding it within their reach. When the recipient did not reach for it (because it had previously made an unsuccessful attempt to obtain the stick), the teaser started waving the object in front of the recipient's face. Only once the recipient reached for the stick, did the teaser withdraw the offer (figure 1d–f). While more systematic observations of object-teasing are needed, this behavioural sequence (waiting for a response and modifying the signal when the response did not occur) suggests that apes produced this offer gesture intentionally to elicit the other's attempt to retrieve the item, a response which they then thwarted by withdrawing the offer. It is undeniably difficult to attribute specific goals to teasers (or gesturers) without relying solely on the intuition of the observer. However, careful examination of the satisfying conditions under which the teaser (or gesturer) stops acting can be used to test the observer's attributions of the goal. This has been a very successful method for analysing the meanings of ape gestures (e.g. [55]). Once the offer is withdrawn, the recipient needs to interpret the intention of the teaser as being affiliative (or neutral) rather than aggressive. Primates typically respond with anger when humans retract offers (e.g. [83]). In order to maintain a positive interaction surrounding the teasing behaviour, recipients cannot rely on the teasing behaviour alone but must take into account their relationship with the teaser, the teaser's affective state and other contextual information. These cognitive inferences are even more critical in the absence of overt play signals (e.g. play-face). Because teasing can be a highly ambiguous behaviour, responding to teasing as play—especially participating in teasing ‘games’ like repeated offers with withdrawal—requires careful assessment of social cues and relationships as well as inferences about the other's motivation in a given interaction (also see [58] for a valuable discussion on how different animal species manage and overcome the ambiguity of actions during play fighting).

5. Humorous play with others' minds?

An intriguing question is whether ape teasers not only expect a specific action response from the recipient but whether they also attribute expectations to their recipient (e.g. the expectation that the teaser will transfer an object). As mentioned above, (false) belief attribution has only been demonstrated recently in apes in a single study employing implicit measures [76]. The occurrence of teasing with offer and withdrawal could provide a hint that apes not only have an implicit understanding of others’ beliefs but that they may even actively create false beliefs by intentionally evoking expectations in the other, before disrupting them.

The deliberate creation of false expectations has previously been discussed in the context of a structurally similar but functionally different behaviour displayed by non-human primates: tactical deception (see [84,85] for reviews). Tactical deception describes ‘acts from the normal repertoire of the agent, deployed such that another individual is likely to misinterpret what the acts signify, to the advantage of the agent’ [81]. There is evidence that great apes use tactical deception in naturally occurring situations [85] and experimental contexts (e.g. [86,87]). Hence, apes do occasionally use false communicative signals to influence the behaviour of others.

One idiosyncrasy of playful teasing is that, in contrast with tactical deception, no immediate fitness benefits are apparent. One possible explanation is that playful teasing constitutes a safe domain within which to explore social rules and boundaries (see [21,2325]). Research on play fighting in apes suggests that individuals can test social rules in play that they might not be able to explore outside the play context (e.g. [88]). Another possibility is that the teasing behaviour evokes a positive affective state in the teaser and perhaps also in the recipient. For human infants, a suggested proximate function of playful teasing is to create a shared humorous experience between teaser and recipient (e.g. [21]), which may strengthen their social bond (but see [27] for an alternative proposition). Social bonds are critically important for fitness in non-human primates [89,90]. Is it, thus, possible that apes also experience positive emotions such as amusement when playfully teasing others and that sharing such moments enhances bonding between individuals?

This question is related to a more general discussion about whether great apes, or any non-human animals, appreciate humour. One widely used definition of humour states that incongruity with respect to reality is the source of humour [28,29]. This incongruity must be in the form of a benign (i.e. harmless) expectation violation; otherwise, it will elicit negative emotions instead of amusement [91]. Creating this incongruity involves a cognitive understanding of action norms and how those can be violated [26]. Great apes have previously demonstrated such understanding in the context of imitation recognition [92,93]. Moreover, apes' playful teasing fulfils the criteria of the benign expectation violation theory [91]. Hence, technically, playful teasing might be viewed as a humorous act. The question is whether apes, like humans, also appreciate this humorous component and experience a positive emotional state during teasing interactions.

In humans, studying humour and its effects on affective states is eased by the fact that, from early infancy onwards, amusement is often (but not always) accompanied by a distinct emotional expression: laughter [94,95]. Importantly, great apes also emit laughter-like vocalizations (though mostly during dynamic social activities like wrestling, tickling and chasing games [9698]), suggesting that apes may experience joy during social interactions. Chimpanzees not only laugh spontaneously but also after hearing the laughter of others [99]. Chimpanzee play sessions involving laughter contagion last longer than play involving only spontaneous laughter (or no laughter at all), suggesting that, like in humans, shared laughter may facilitate positive social interaction and enhance bonding (also see [100,101] for evidence of contagious play vocalizations in rats and kea parrots).

Studies documenting offer-withdrawal, provocative non-compliance or disruption of other's activities in apes reported that these behaviours occurred in playful contexts and, thus, likely involved a positive emotional state. However, most studies did not report on any affective signals, such as play-face or laughter (but see [48,50]). Hence, while teasing constitutes an excellent place to look for potential antecedents of joking behaviour and humour in great apes, future research will need to pay close attention to markers of positive affect during these activities. Finding evidence that both teaser and recipient exhibit positive affective states would strengthen the hypothesis that non-human animals are capable of creating and appreciating humorous experiences, and that they, like human infants, use mild expectation-violations to strengthen their bonds.

Psychological Profile of Extreme Trump Supporters: Many of these tendencies generalize to extreme Trump opposition, meaning that Trump support is in many ways an extrapolation of a long-standing and escalating Culture War

A Psychological Profile of Extreme Trump Supporters. Laura Kinsman & Jeremy A. Frimer. Sep 2020. http://www.jeremyfrimer.com/uploads/2/1/2/7/21278832/kinsman___frimer__in_press_.pdf

Abstract: Donald Trump once claimed that he could murder someone in broad daylight and still not lose support, implying an extreme loyalty on the part of his supporters. The psychological foundation of this purported commitment is far from obvious in that many aspects of Trump’s lifestyle, personality, policies, and values are antithetical to those cherished by large segments of his base (e.g., Evangelical Christians, free trade advocates). Political scholars, pundits and the media have offered a number of explanations for the support for Trump that exists within his base. In this chapter, we review three of the more common media explanations, which include tribal loyalties, selective media exposure, and material self-interest, and find some support for each in the scientific literature, with support generally being weaker for the more nefarious explanation (e.g., tribal racism, bloodlust) and stronger for more banal explanations (e.g., information bubbles, perceived self-interest). We also find that many of these tendencies generalize to extreme Trump opposition, meaning that Trump support is in many ways an extrapolation of a long-standing and escalating Culture War.



Thanksgiving Dinners: Politics is not straining family ties as much as previously thought

Are Politically Diverse Thanksgiving Dinners Shorter Than Politically Uniform Ones? In press in PLOS One. Jeremy A. Frimer & Linda J. Skitka. http://www.jeremyfrimer.com/uploads/2/1/2/7/21278832/frimer_skitka_in_press_.pdf

Abstract: Americans on the political left and right are engaged in a Culture War with one another, one that is often characterized by mutual fear, antipathy, and avoidance. Are there safe havens from the socially straining effects of this Culture War, times and places where Americans of different political stripes gather and put aside their political differences? Previous research (Chen & Rohla, 2018) implied that there might not be insofar as even intimate family gatherings seem to be subject to Culture War tensions. They found that politically diverse Thanksgiving Dinners were 35-70 minutes shorter than politically uniform ones, representing a 14-27% reduction in overall dinner duration. Noting analytical and methodological limitations in the prior analysis, we conducted two pre-registered studies to test whether diverse dinners are shorter than uniform ones and to attempt to conceptually replicate and extend this prior analysis. Individual analyses yielded mixed results, with null models generally supported but effect estimates generally overlapping with those of Chen and Rohla (2018). A mega-analysis found that, when controlling for various covariates, politically diverse dinners were 24 minutes shorter than politically uniform ones, 95% confidence interval = [9, 39], representing a 6% decrease in the total dinner time [2%-10%]. This final result successfully replicates Chen and Rohla (2018) both in terms of effect overlap and direct-and-significance criteria while nonetheless favoring the conclusion that politics is not straining family ties as much as previously thought.

Keywords: liberals and conservatives, intergroup relations, Culture War, Thanksgiving, selective exposure