Saturday, March 27, 2021

The most human bot: Female gendering increases humanness perceptions of bots and acceptance of AI

The most human bot: Female gendering increases humanness perceptions of bots and acceptance of AI. Sylvie Borau  Tobias Otterbring  Sandra Laporte  Samuel Fosso Wamba. Psychology & Marketing, March 22 2021.

Abstract: Companies have repeatedly launched Artificial Intelligence (AI) products such as intelligent chatbots and robots with female names, voices, and bodies. Previous research posits that people intuitively favor female over male bots, mainly because female bots are judged as warmer and more likely to experience emotions. We present five online studies, including four preregistered, with a total sample of over 3,000 participants that go beyond this longstanding perception of femininity. Because warmth and experience (but not competence) are seen as fundamental qualities to be a full human but are lacking in machines, we argue that people prefer female bots because they are perceived as more human than male bots. Using implicit, subtle, and blatant scales of humanness, our results consistently show that women (Studies 1A and 1B), female bots (Studies 2 and 3), and female chatbots (Study 4) are perceived as more human than their male counterparts when compared with non‐human entities (animals and machines). Study 4 investigates explicitly the acceptance of gendered algorithms operated by AI chatbots in a health context. We found that the female chatbot is preferred over the male chatbot because it is perceived as more human and more likely to consider our unique needs. These results highlight the ethical quandary faced by AI designers and policymakers: Women are said to be transformed into objects in AI, but injecting women's humanity into AI objects makes these objects seem more human and acceptable.

People tend to naturally drop their efforts to be optimistic when they expect things to go badly, especially when being evaluated; people likely know that optimism is not the most beneficial mindset to adopt at all times

Optimism: Enduring resource or miscalibrated perception? Mariah F. Purol  William J. Chopik. Social and Personality Psychology Compass, March 25 2021.

Abstract: There is a general, widely‐held belief that optimism is always a good thing. While there is much previous research suggesting that optimists enjoy several health and wellness benefits, there is also a large body of research suggesting that optimism is not always advantageous. Perhaps examining how optimism develops and changes across the lifespan may give us insight into how people use optimism and allow us to determine if and when optimism is helpful or maladaptive for them. In the current review, we review evidence debating the benefits and costs of optimism, as well as examine how optimism develops across the lifespan. We discuss how life events may or may not impact the developmental trajectory of optimism. Lastly, we address currently unanswered questions and emphasize the contextual nature of optimism's advantages.


Of course, there are situations where optimism can do more harm than good. Take, for example, what researchers have coined as unrealistic optimism. Unrealistic optimism is the belief that one is more likely to experience positive outcomes compared to others who are objectively similar to them (Weinstein, 1980). In an often‐cited example of unrealistic optimism, smokers believe that they are at less risk for developing lung cancer compared to the general population of smokers (Weinstein et al., 2005).

This mindset comes with some clearly negative implications for health behaviors. For smokers, it leads to a discounting of a very real health risk and might interfere with efforts to quit smoking. While some research has suggested that optimists may be more attentive to information about potential risks (Aspinwall & Brunhart, 1996), those who are high in unrealistic optimism may avoid this same information (Wiebe & Black, 1997), which may ultimately stop them from fully understanding their risk or acting preventatively.

Importantly and, perhaps, counterintuitively, unrealistic optimism is often assessed independently of accuracy (Weinstein & Klein, 1996). Thus, it is difficult to determine if one is truly unrealistic when they say that they have a lower risk of developing any given health condition than the average person; maybe people who make these kinds of claims do indeed have more positive outcomes and are different from others in a similar boat. However, in studies that evaluate the chances of a specific outcome (e.g., evaluate the risk of heart attack using blood pressure and cholesterol data), researchers have operationalized when optimism is considered “unrealistic” (e.g., those who misjudged their risk by greater than 10%; Radcliffe & Klein, 2002). This research has found that, while dispositional optimists have a lower risk of negative outcomes, unrealistic optimists have a higher risk of negative outcomes.

Unrealistic optimists are less worried about their risk levels for negative events (Weinstein, 1982), have less prior knowledge about risks, and remember less when provided with information about risk (Radcliffe & Klein, 2002). Some researchers have suggested that this may be because of the invulnerability sometimes felt by unrealistic optimists (Perloff & Fetzer, 1986; Schwarzer, 1994); they may feel that risk information is irrelevant to them. Weinstein and Lachendro (1982) suggest that egocentrism plays a role in our use of unrealistic optimism—we tend to think that we will be far better off than others when we are not forced to think carefully about others' circumstances. Neuroscience research on the topic has suggested that those high in unrealistic optimism fail to code for errors that should reduce optimism, making it difficult for them to accurately update their beliefs (Sharot et al., 2011). Altogether, an unrealistic sense of optimism leads people to be at higher risk for negative outcomes, seek out less information about risk, and take fewer preventative steps to mitigate risk. Of course, unrealistic optimism is not a dichotomy, and those who are very high in unrealistic optimism may be most at risk for these aversive outcomes.

In the past decade, many researchers have found that optimism, even when not unrealistic, can occasionally be associated with negative outcomes. In another often‐cited example, being optimistic about exam scores does not make students feel any less distressed or nervous before they get feedback, and being optimistic does not protect students from feeling bad when they learn they did poorly (Sweeny & Shepperd, 2010). In fact, optimism leads to greater disappointment when students receive a bad grade. Interestingly, students know that getting their hopes up, only to receive a bad grade, will be disappointing—and, yet, they continue to be optimistic (Sweeny & Shepperd, 2010). This suggests that, despite knowing the costs of optimism and experiencing no positive change in affect because of it, students continue to be optimistic. Worth noting, true pessimists—those who believe that they performed worse on the exam than they really did—report lower negative affect after the feedback (Sweeny & Shepperd, 2010).

Further, people tend to naturally drop their efforts to be optimistic when they expect things to go badly, especially when being evaluated. This suggests that people likely know that optimism is not the most beneficial mindset to adopt at all times. This tendency to shift towards pessimism in the moments before feedback is referred to as “sobering up” (Sweeny & Krizan, 2013). In general, the closer we get to an evaluation of our performance, the more pessimistic we become. There are many reasons why this may occur. For example, the closer we get to an event, the less control we have over the outcome, and the more “real” (i.e., concrete) it becomes (Sweeny & Krizan, 2013). There is an increased pressure to be accurate in our prediction of how we will fare after an event, and we are more likely to think critically about our expectations in order to counteract any unrealistic optimism (Lerner & Tetlock, 1999; Tetlock & Kim, 1987). As in the case of the student receiving their exam score, a shift toward pessimism may also be an outcome of affect management, in which we temper our expectations in order to avoid negative feelings (Sweeny & Krizan, 2013). Being pessimistic in these moments can spare us from painful emotions, like disappointment, and allow us to prepare for unfavorable outcomes (Sweeny et al., 2006). The protective functions of bracing for bad news may explain why, in some cases, pessimism might be called for—and that optimism might be a bad thing.


When determining the efficacy of optimism, it is important to examine the quality of evidence. How convincing is current research?

There are reasons to be skeptical—many studies on the benefits of optimism rely on correlational data, including many of those discussed above (i.e., Andersson; 1996; Carver et al., 1989; Gould et al., 2002; Nes & Segerstrom, 2006; Scheier & Carver, 1992). Anderson (1996) goes as far as to note that “practically all studies of the benefits of optimism as assessed by the LOT [the Life Orientation Test, a popular optimism measure] have been in the form of correlational designs.” However, more recent work has applied more rigorous methods of analysis.

Interventions, for example, offer an opportunity to examine if optimism can be manipulated and test its connection to specific outcomes. In a meta‐analysis of interventions, Malouff and Schutte (2017) determined that, while these programs are, overall, successful in increasing optimism, this success may be highly dependent on methodology. Timing of measurement, instrument used, intervention length, and other methodological artifacts were moderators of effect sizes yielded from the interventions (Malouff & Schutte, 2017). Of course, all interventions are not created equal. Some interventions, like the “best possible selves” exercise—in which participants imagine themselves in the best possible future and what they have done to get there—have been successful in both boosting optimism itself and in using optimism to increase positive affect (Carrillo et al., 2019; Malouff & Schutte, 2017). Other interventions, such as cognitive‐behavioral techniques, have also found success in cultivating long‐term gains in optimism (Brunwasser et al., 2009).

In a recent meta‐analysis of optimism's associations with positive health behaviors, Boehm et al. (2018) identified other common pitfalls of optimism research, including its reliance on cross‐sectional research. Indeed, much of the work discussed above, arguing both for and against optimism, is cross sectional. However, the longitudinal work that does exist, much of which examines the entire adult lifespan (Chopik et al., 20152018; Daukantaite & Bergman, 2005; Daukantaitė & Zukauskiene, 2012; Kim et al., 2014), suggests these findings are likely not just an artifact of cross‐sectional analyses. In general, less longitudinal work has examined the costs of optimism; the few existing exceptions have suggested that, without intervention, unrealistically optimistic individuals may be at risk for poorer cognitive, performance, and health outcomes than their more realistic counterparts (Haynes et al., 2006; Popova & Halpern‐Felsher, 2016).

With this in mind, there are other pieces of evidence that offer insight into the efficacy of optimism. Longitudinal work on how optimism changes throughout the lifespan offers one such insight.

Why sexual coercion is so prevalent in orangutans & how this type of sexual selection may be much more common across animals than often recognized

Orangutan socio-sexual behavior and sexual conflict: Insights for human evolution. Cheryl D. Knott. European Human Behaviour and Evolution Association, 15th Conference, Mar 2021.

Abstract: In this talk I reveal how recent research on great ape behavior and physiology provides new insights into the similarities we share with our closest relatives. In particular, I focus on my long-term research studying wild orangutans in Gunung Palung National Park, Indonesia for over 25 years. Orangutans are known for one of the highest rates of sexual coercion, through forced copulation, of any animal. This is coupled with another intriguing phenomenon of having two male morphs, a rare type of male bi-maturism. Females share crucial features of reproductive physiology in common with humans, such as concealed ovulation and menstrual cycle length. In this talk I explore the complexity of male and female reproductive decisions in wild orangutans and the ways that these reveal insights into the evolution of human mating systems. This includes new research from my team on the development of socio-sexual behavior in adolescent females and how the threat of forced copulation, as well possible infanticide risk, impacts female behavior and ranging patterns. I also demonstrate the success of strategies employed by females to avoid undesired sires. These results reveal that, despite high rates of forced copulation, female choice is an important feature of orangutan mating patterns. I also discuss why sexual coercion is so prevalent in orangutans and how this type of sexual selection may be much more common across animals than often recognized. I point to the need for considering comparative data on sexual conflict as we consider the evolution of human mating patterns.

Our work shows that maternal childhood trauma can affect infant growth parameters; it also suggests that early maternal stress might set child’s development for faster life trajectory

Accelerated growth in infants of mothers with early childhood trauma. Apanasewicz-Grzegorczyk, A; Danel, D; Ziomkiewicz-Wichary, A. European Human Behaviour and Evolution Association, 15th Conference, Mar 2021.

Abstract: Early life stress has long-term programming effect on growth, development and further health. Recent experimental studies in animals demonstrated that the effect of early maternal psychological stress may extend on the offspring. The aim of the presented study is to investigate the effect of early maternal trauma on infant growth parameters. To test this effect, we studied a sample of 99 exclusively breastfeeding mothers and their healthy, born on-time, 5 months old infants. Mothers were asked to complete Early Life Stress Questionnaire (ELSQ) to assess maternal trauma during childhood. Anthropometrical measurements of infant body length, weight and head circumference were taken. Multivariate analysis of variance (MANCOVA) was used to test for the effect of maternal trauma and infant sex on infant’s growth parameters. Maternal childhood trauma and infant sex significantly predicted infant growth parameters. However, no effect of interaction between maternal trauma and infant sex was detected. The maternal childhood trauma was positively associated with infant growth parameters (λ =.90, F (3,93) = 3. 42, p = .02,η2 = .10). The separate univariate models indicated trauma and infant sex effect on body weight and head circumference and infant sex effect on body length. Our results demonstrate that maternal childhood trauma can affect infant growth parameters. They also suggest that early maternal stress might set child’s development for faster life trajectory.

Republicans have a higher within-party facial resemblance than Democrats; UK Conservative MPs are more similar looking to each other than Labour

Pahontu, Raluca L. and Poupakis, Stavros, Resemblance and Discrimination in Elections (March 22, 2021). SSRN:

Abstract: Discrimination affects hiring, mating and voting decisions. Whilst discrimination in elections mainly relates to gender or race, we introduce a novel source of discrimination: candidate resemblance. When candidates' partisanship is not known, voters select those that resemble most elected co-partisans. Using a machine learning algorithm for face comparison, we find a stronger resemblance effect for Republicans compared to Democrats in the US. This happens because Republicans have a higher within-party facial resemblance than Democrats, even when accounting for gender and race. We find a similar pattern in the UK, where Conservative MPs are more similar looking to each other than Labour. Using a survey experiment, we find that Tory voters reward resemblance, while there is no similar effect for Labour. We estimate that facial dissimilarity decreases the candidate's re-election probability by 5-14 percentage points. The results are consistent with an interpretation of this behaviour as a form of statistical discrimination.

Keywords: Voter Behaviour, Discrimination, Facial Resemblance, Low-Information Election, Partisanship

JEL Classification: D72, D83

Individuals high on Machiavellianism or psychopathy traits choose to stay single if they are low on sociosexuality

Individuals high on the Dark Triad traits choose to stay single if they are low on sociosexuality. Vlad Burtăverde, Cristina Ene. Personality and Individual Differences, Volume 177, July 2021, 110843.

Abstract: Even if the majority of humans desire to mate, some people decide to stay single. In this paper, we investigated (N = 270) the link between the Dark Triad traits and the reasons to remain single by choice, testing the moderating effect of sociosexuality. We showed that individuals high on Machiavellianism and psychopathy scored high on two dimensions of reasons to stay single by choice: freedom of choice and difficulties with relationships. Narcissism was not related to any dimension of reasons to stay single. Individuals high on Machiavellianism or psychopathy that were also high on sociosexuality reported lower scores on the reasons to stay single by choice, compared to individuals high on Machiavellianism or psychopathy low on sociosexuality.

Keywords: Dark triad traitsMating behaviorSinglehoodSociosexuality

Our pattern of pelvic sex differences did not evolve de novo in modern humans and must have been present in the common ancestor of humans and chimpanzees, and thus also in the extinct Homo species

Sex differences in the pelvis did not evolve de novo in modern humans. Barbara Fischer, Nicole D. S. Grunstra, Eva Zaffarini & Philipp Mitteroecker. Nature Ecology & Evolution, Mar 25 2021.

Abstract: It is commonly assumed that the strong sexual dimorphism of the human pelvis evolved for delivering the relatively large human foetuses. Here we compare pelvic sex differences across modern humans and chimpanzees using a comprehensive geometric morphometric approach. Even though the magnitude of sex differences in pelvis shape was two times larger in humans than in chimpanzees, we found that the pattern is almost identical in the two species. We conclude that this pattern of pelvic sex differences did not evolve de novo in modern humans and must have been present in the common ancestor of humans and chimpanzees, and thus also in the extinct Homo species. We further suggest that this shared pattern was already present in early mammals and propose a hypothesis of facilitated variation as an explanation: the conserved mammalian endocrine system strongly constrains the evolution of the pattern of pelvic differences but enables rapid evolutionary change of the magnitude of sexual dimorphism, which in turn facilitated the rapid increase in hominin brain size.

Likelihood of overweight/obesity was higher among Nepalese males compared with females who used the internet frequently

Sex Differences in the Association between Internet Usage and Overweight/Obesity: Evidence from a Nationally Representative Survey in Nepal. Juwel Rana, Momin Islam, John Oldroyd, Nandeeta Samad, Rakibul Islam. Sexes 2021, 2(1), 132-143; March 18 2021.


Objective: To examine the associations between internet use and overweight/obesity in people aged 15–49 years in Nepal and the extent to which these associations differ by biological sex.

Materials and methods: The study analyzed the nationally representative Nepal Demographic and Health Survey (NDHS) 2016 data. Multivariable ordinal logistic regression models were fitted to estimate the total effects of internet use (IU) in the last 12 months and frequency of internet use (FIU) in the last month on overweight/obesity adjusted for potential confounders.

Results: Of the 10,380 participants, the prevalence of overweight/obesity by IU was 38% (95% confidence interval (CI): 35.9%, 40.1%) for males and 44.1% (95% CI: 41.6%, 46.6%) for female. The likelihood of overweight/obesity was significantly higher (adjusted odds ratio (aOR): 1.55; 95% CI: 1.40, 1.73; p < 0.001) among those participants who used the internet compared to the participants who did not use the internet in the last 12 months. Similar associations were observed when using the augmented measure of exposure-FIU in the last month. We observed the modification effect of sex in the associations of IU (p-difference < 0.001) and FIU (p-difference < 0.002) with overweight/obesity in Nepal.

Conclusions: Our findings suggest that future overweight/obesity interventions in Nepal discourage unnecessary internet use, particularly among males.

Keywords: sex; internet; obesity; overweight; sex; sex differences; Nepal

Social preference in rats: They preferred social over nonsocial options, choosing their cagemate rat over an empty chamber, and an unfamiliar over a familiar rat, choosing a non‐cagemate over their cagemate

Social preference in rats. Timothy D. Hackenberg  Lauren Vanderhooft  Jasmine Huang  Madeline Wagar  Jordan Alexander  Lavinia Tan. Journal of the Experimental Analysis of Behavior, March 13 2021

Rolf Degen's take:

Abstract: Rats were given repeated choices between social and nonsocial outcomes, and between familiar and unfamiliar social outcomes. Lever presses on either of 2 levers in the middle chamber of a 3‐chamber apparatus opened a door adjacent to the lever, permitting 45‐s access to social interaction with the rat in the chosen side chamber. In Experiment 1, rats preferred (a) social over nonsocial options, choosing their cagemate rat over an empty chamber, and (b) an unfamiliar over a familiar rat, choosing a non‐cagemate over their cagemate. These findings were replicated in Experiment 2 with 2 different non‐cagemate rats. Rats preferred both non‐cagemate rats to a similar degree when pitted against their cagemate, but were indifferent when the 2 non‐cagemates were pitted against each other. Similar preference for social over nonsocial and non‐cagemate over cagemate was seen in Experiment 3, with new non‐cagemate rats introduced after every third session. Response rates (for both cagemate and non‐cagemate rats) were elevated under conditions of nonsocial (isolated) housing compared to conditions of social (paired) housing, demonstrating a social deprivation effect. Together, the experiments contribute to an experimental analysis of social preference within a social reinforcement framework, drawing on methods with proven efficacy in the analysis of reinforcement more generally.


I knew I was a rat...

Harm inflation: Concepts creep to the Left and the Right

Harper, Craig A., Harry R. M. Purser, and Thom Baguley. 2021. “Concepts Creep to the Left and the Right.” PsyArXiv. March 26. doi:10.31234/

Abstract: In a target article in 2017, social psychologist Nick Haslam proposed that concept creep explains how established social concepts expand to incorporate new phenomena, with such expansions fundamentally changing conceptual definitions and contributing to a loss of a shared social understanding. However, Haslam’s piece (along with several commentaries) focused on concept creep in relation to a small number of categories (e.g., prejudice, bullying, trauma) that are typically more salient for those on the political left. In this work, we examined whether concept creep is a uniquely leftist phenomenon, or whether we can observe the same conceptual expansion for categories typically salient for conservatives. We found evidence for such symmetry when considering categories such as sexual deviance, terrorism, and personal responsibility – with some nuanced exceptions. We discuss our findings in relation to growing political polarization, intergroup relations, and the study of partisan differences using a variety of politically salient stimuli.

Check also... When objective harm decreases, concepts of harm may expand to encompass new & previously innocuous phenomena, so it appears as widespread as ever; sometimes via intentional meaning changes engineered for political ends

Harm inflation: Making sense of concept creep. Nick Haslam et al. European Review of Social Psychology, Volume 31, 2020 - Issue 1, Jul 22 2020.

The administration of testosterone to women caused a significant enlargement of participants’ peripersonal space, suggesting that testosterone caused participants to implicitly appropriate a larger space as their own

Testosterone administration in women increases the size of their peripersonal space. Catherine Masson, Donné van der Westhuizen, Jean-Paul Noel, Adala Prevost, Jack van Honk, Aikaterini Fotopoulou, Mark Solms & Andrea Serino. Experimental Brain Research, Mar 26 2021.

Abstract: Peripersonal space (PPS) is the space immediately surrounding the body, conceptualised as a sensory-motor interface between body and environment. PPS size differs between individuals and contexts, with intrapersonal traits and states, as well as social factors having a determining role on the size of PPS. Testosterone plays an important role in regulating social-motivational behaviour and is known to enhance dominance motivation in an implicit and unconscious manner. We investigated whether the dominance-enhancing effects of testosterone reflect as changes in the representation of PPS in a within-subjects testosterone administration study in women (N = 19). Participants performed a visuo-tactile integration task in a mixed-reality setup. Results indicated that the administration of testosterone caused a significant enlargement of participants’ PPS, suggesting that testosterone caused participants to implicitly appropriate a larger space as their own. These findings suggest that the dominance-enhancing effects of testosterone reflect at the level of sensory-motor processing in PPS.