Never tell me the odds
Robots that can adapt like animals
Antoine Cully et al.
Nature, 28 May 2015, Pages 503–507
Abstract:
Robots have transformed many industries, most notably manufacturing, and have the power to deliver tremendous benefits to society, such as in search and rescue, disaster response, health care and transportation. They are also invaluable tools for scientific exploration in environments inaccessible to humans, from distant planets to deep oceans. A major obstacle to their widespread adoption in more complex environments outside factories is their fragility. Whereas animals can quickly adapt to injuries, current robots cannot 'think outside the box' to find a compensatory behaviour when they are damaged: they are limited to their pre-specified self-sensing abilities, can diagnose only anticipated failure modes, and require a pre-programmed contingency plan for every type of potential damage, an impracticality for complex robots. A promising approach to reducing robot fragility involves having robots learn appropriate behaviours in response to damage, but current techniques are slow even with small, constrained search spaces. Here we introduce an intelligent trial-and-error algorithm that allows robots to adapt to damage in less than two minutes in large search spaces without requiring self-diagnosis or pre-specified contingency plans. Before the robot is deployed, it uses a novel technique to create a detailed map of the space of high-performing behaviours. This map represents the robot's prior knowledge about what behaviours it can perform and their value. When the robot is damaged, it uses this prior knowledge to guide a trial-and-error learning algorithm that conducts intelligent experiments to rapidly discover a behaviour that compensates for the damage. Experiments reveal successful adaptations for a legged robot injured in five different ways, including damaged, broken, and missing legs, and for a robotic arm with joints broken in 14 different ways. This new algorithm will enable more robust, effective, autonomous robots, and may shed light on the principles that animals use to adapt to injury.
---------------------
Cognitive capacities for cooking in chimpanzees
Felix Warneken & Alexandra Rosati
Proceedings of the Royal Society: Biological Sciences, 22 June 2015
Abstract:
The transition to a cooked diet represents an important shift in human ecology and evolution. Cooking requires a set of sophisticated cognitive abilities, including causal reasoning, self-control and anticipatory planning. Do humans uniquely possess the cognitive capacities needed to cook food? We address whether one of humans' closest relatives, chimpanzees (Pan troglodytes), possess the domain-general cognitive skills needed to cook. Across nine studies, we show that chimpanzees: (i) prefer cooked foods; (ii) comprehend the transformation of raw food that occurs when cooking, and generalize this causal understanding to new contexts; (iii) will pay temporal costs to acquire cooked foods; (iv) are willing to actively give up possession of raw foods in order to transform them; and (v) can transport raw food as well as save their raw food in anticipation of future opportunities to cook. Together, our results indicate that several of the fundamental psychological abilities necessary to engage in cooking may have been shared with the last common ancestor of apes and humans, predating the control of fire.
---------------------
Page Moreau & Marit Gundersen Engeset
Journal of Marketing Research, forthcoming
Abstract:
Business leaders, governments, and scholars are increasingly recognizing the importance of creativity. Recent trends in technology and education, however, suggest that many individuals are facing fewer opportunities to engage in creative thought as they increasingly solve well-defined (versus ill-defined) problems. Using three studies that involve real problem-solving activities (e.g., putting together a Lego kit), the authors examine the mindset created by addressing such well-defined problems. The studies demonstrate the negative downstream impact of such a mindset on both creative task performance and the choice to engage in creative tasks. The research has theoretical implications for the creativity and mindset literatures as well as substantive insights for managers and public-policy makers.
---------------------
Knowing More Than We Can Tell: People Are Aware of Their Biased Self-Perceptions
Kathryn Bollich, Katherine Rogers & Simine Vazire
Personality and Social Psychology Bulletin, July 2015, Pages 918-929
Abstract:
There is no question that biases exist in self-perceptions of personality. To what extent do people have insight into their positive and negative self-biases? In two samples (total N = 130), people with positive biases (i.e., self-perceptions that are more positive than a reputation-based criterion measure) accurately described themselves as positively biased, and people with negative biases accurately described themselves as negatively biased. Furthermore, people were able to distinguish which traits they were more or less biased about. These findings suggest that people may know more about themselves than they initially admit. Implications for the use of self-reports and the study of self-knowledge are discussed.
---------------------
The Effect of Foreign Language in Judgments of Risk and Benefit: The Role of Affect
Constantinos Hadjichristidis, Janet Geipel & Lucia Savadori
Journal of Experimental Psychology: Applied, June 2015, Pages 117-129
Abstract:
As a result of globalization, policymakers and citizens are increasingly communicating in foreign languages. This article investigates whether communicating in a foreign language influences lay judgments of risk and benefit regarding specific hazards such as "traveling by airplane," "climate change," and "biotechnology." Merging findings from bilingual and risk perception research, we hypothesized that stimuli described in a foreign language, as opposed to the native tongue, would prompt more positive overall affect and through that induce lower judgments of risk and higher judgments of benefit. Two studies support this foreign language hypothesis. Contrary to recent proposals that foreign language influences judgment by promoting deliberate processing, we show that it can also influence judgment through emotional processing. The present findings carry implications for international policy, such as United Nations decisions on environmental issues.
---------------------
Rajesh Bagchi & Elise Chandon Ince
Journal of Marketing Research, forthcoming
Abstract:
Consumers routinely rely on forecasters to make predictions about uncertain events (e.g., sporting contests, stock fluctuations). The authors demonstrate that when forecasts are higher versus lower (e.g., a 70% vs. 30% chance of team A winning a game) consumers infer that the forecaster is more confident in her prediction, has conducted more in-depth analyses, and is more trustworthy. The prediction is also judged as more accurate. This occurs because forecasts are evaluated based on how well they predict the target event occurring (team A winning). Higher forecasts indicate greater likelihood of the target event occurring, and signal a confident analyst, while lower forecasts indicate lower likelihood and lower confidence in the target event occurring. But because, with lower forecasts, consumers still focus on the target event (and not its complement), lower confidence in the target event occurring is erroneously interpreted as the forecaster being less confident in her overall prediction (instead of more confident in the complementary event occurring — team A losing). The authors identify boundary conditions, generalize to other prediction formats, and demonstrate consequences.
---------------------
Learning by Thinking: Overcoming the Bias for Action through Reflection
Giada Di Stefano et al.
Harvard Working Paper, March 2015
Abstract:
Research on learning has primarily focused on the role of doing (experience) in fostering progress over time. Drawing on literature in cognitive psychology and neuroscience, we propose that one of the critical components of learning is reflection, or the intentional attempt to synthesize, abstract, and articulate the key lessons taught by experience. In particular, we argue that purposeful reflection on one's accumulated experience leads to greater learning than the accumulation of additional experience. We explain this boost in learning through self-efficacy: reflection builds confidence in the ability to achieve a goal, which in turn translates into higher rates of learning. We test the resulting model experimentally, using a mixed-method design that combines two laboratory experiments with a field experiment conducted in a large business-process outsourcing company in India. We find that individuals who are given time to reflect on a task improve their performance at a greater rate than those who are given the same amount of time to practice with the same task. Our results also show that if individuals themselves are given the choice to either reflect or practice, they prefer to allocate their time to gaining more experience with the task – to the detriment of their learning.
---------------------
Group discussion improves lie detection
Nadav Klein & Nicholas Epley
Proceedings of the National Academy of Sciences, forthcoming
Abstract:
Groups of individuals can sometimes make more accurate judgments than the average individual could make alone. We tested whether this group advantage extends to lie detection, an exceptionally challenging judgment with accuracy rates rarely exceeding chance. In four experiments, we find that groups are consistently more accurate than individuals in distinguishing truths from lies, an effect that comes primarily from an increased ability to correctly identify when a person is lying. These experiments demonstrate that the group advantage in lie detection comes through the process of group discussion, and is not a product of aggregating individual opinions (a "wisdom-of-crowds" effect) or of altering response biases (such as reducing the "truth bias"). Interventions to improve lie detection typically focus on improving individual judgment, a costly and generally ineffective endeavor. Our findings suggest a cheap and simple synergistic approach of enabling group discussion before rendering a judgment.
---------------------
Krishna Savani & Dan King
Organizational Behavior and Human Decision Processes, forthcoming
Abstract:
People view the same decision as better when it is followed by a positive outcome than by a negative outcome, a phenomenon called the outcome bias. Based on the idea that a key cause of the outcome bias is people's failure to appreciate that outcomes are in part determined by external forces, three studies tested a novel method to reduce the outcome bias. Experiment 1 showed that people who construed a person's interactions with the environment as events rather than as actions or choices were less susceptible to the outcome bias in a medical decision making task. Experiments 2 and 3 demonstrated that people who recalled past events rather than actions or choices exhibited lower outcome bias in a risky decision making task and in an ethical judgment task. These findings indicate that an event construal helps people appreciate the role of external factors in causing outcomes.
---------------------
The Impact of a Relational Mindset on Information Distortion
Anne-Sophie Chaxel
Journal of Experimental Social Psychology, September 2015, Pages 1–7
Abstract:
The preference-supporting bias in information evaluation, known as information distortion, is a ubiquitous phenomenon. The present work demonstrates that priming a relational mindset induces individuals to process independent units of information interdependently and therefore contributes to increasing distortion. In three studies, a relational mindset is activated by asking participants to generate solutions to cross-domain analogies. All three studies show that the activation of a relational mindset then carries over into a second, unrelated choice task and increases distortion. In addition, the present work shows that generating solutions to cross-domain analogies activates a high level of construal, which in turn mediates the effect of relational thinking on information distortion. Finally, the present work also demonstrates that imposing a cognitive load during the choice task reduces the impact of the relational mindset on distortion. In sum, this research demonstrates that the same mechanism that promotes creative thinking (i.e., seeing relationships across concepts) may also induce more biased information processing by prompting individuals to process independent units of information interdependently.
---------------------
Maferima Touré-Tillery & Ann McGill
Journal of Marketing, forthcoming
Abstract:
Participants in three studies read advertisements in which messages were delivered by people or by anthropomorphized agents, specifically, "talking" products. Results indicate that people low in interpersonal trust are more persuaded by anthropomorphized messengers than by human spokespersons due their greater attentiveness to the nature of the messenger and their belief that humans, more than partial humans (that is, anthropomorphized agents), lack goodwill. People high in interpersonal trust are less attentive about who is trying to persuade them, so respond similarly to human and anthropomorphized messengers. However, when prompted to be attentive, they are more persuaded by human spokespersons than by anthropomorphized messengers, due to their belief that humans, more than partial humans, act with goodwill. Under conditions in which attentiveness is low for all consumers, high and low trusters alike, are unaffected by the nature of persuasion agents. We discuss the implications of our findings for advertisers considering the use of anthropomorphized "spokespersons."
---------------------
Alvaro San Martin et al.
Organizational Behavior and Human Decision Processes, May 2015, Pages 49–60
Abstract:
Two studies examined whether expecting future interaction with the same group members affects minority influence. Holding constant majority members' expectation of future interaction, Study 1 demonstrated that minorities had more influence on majorities' private decisions and the group's public decision when they did not expect future interaction with the majority than when they did. Study 2 demonstrated that this minority influence effect only emerged when majority members themselves expected future interaction. Study 2 also shed light on the early information sharing dynamics underlying this effect: minorities expressed more dissent when they did not expect future interaction and majorities were more open to divergent information when they expected future interaction. These two forces combined promoted more systematic information processing by the group as a whole and, eventually, resulted in greater minority influence on both private and public decisions. Implications for our understanding of minority influence and group decision-making are discussed.
---------------------
Jesse Shore, Ethan Bernstein & David Lazer
Organization Science, forthcoming
Abstract:
Using data from a novel laboratory experiment on complex problem solving in which we varied the structure of 16-person networks, we investigate how an organization's network structure shapes the performance of problem-solving tasks. Problem solving, we argue, involves both exploration for information and exploration for solutions. Our results show that network clustering has opposite effects for these two important and complementary forms of exploration. Dense clustering encourages members of a network to generate more diverse information but discourages them from generating diverse theories; that is, clustering promotes exploration in information space but decreases exploration in solution space. Previous research, generally focusing on only one of those two spaces at a time, has produced an inconsistent understanding of the value of network clustering. By adopting an experimental platform on which information was measured separately from solutions, we bring disparate results under a single theoretical roof and clarify the effects of network clustering on problem-solving behavior and performance. The finding both provides a sharper tool for structuring organizations for knowledge work and reveals challenges inherent in manipulating network structure to enhance performance, as the communication structure that helps one determinant of successful problem solving may harm the other.
---------------------
When Your Decisions Are Not (Quite) Your Own: Action Observation Influences Free Choices
Geoff Cole et al.
PLoS ONE, May 2015
Abstract:
A growing number of studies have begun to assess how the actions of one individual are represented in an observer. Using a variant of an action observation paradigm, four experiments examined whether one person's behaviour can influence the subjective decisions and judgements of another. In Experiment 1, two observers sat adjacent to each other and took turns to freely select and reach to one of two locations. Results showed that participants were less likely to make a response to the same location as their partner. In three further experiments observers were asked to decide which of two familiar products they preferred or which of two faces were most attractive. Results showed that participants were less likely to choose the product or face occupying the location of their partner's previous reaching response. These findings suggest that action observation can influence a range of free choice preferences and decisions. Possible mechanisms through which this influence occurs are discussed.
---------------------
Group brainstorming: When regulatory nonfit enhances performance
John Levine et al.
Group Processes & Intergroup Relations, forthcoming
Abstract:
In a study investigating motivational factors in group brainstorming, we investigated the impact on performance of regulatory fit/nonfit (based on the group's focus and task strategy). All members of three-person groups were placed in either a promotion focus or a prevention focus and then were all given either an eager strategy or a vigilant strategy for performing a brainstorming task (with an expectancy stop rule). As predicted, groups experiencing nonfit (promotion/vigilant, prevention/eager) spent more time working on the task and generated more nonredundant ideas than did groups experiencing fit (promotion/eager, prevention/vigilant). Also as predicted, task persistence mediated the joint impact of regulatory focus and task strategy on idea generation. Parallel results for idea diversity and quality were accounted for by number of ideas generated. These findings shed light on motivational aspects of group brainstorming and demonstrate the utility of regulatory fit theory for explaining small group phenomena.
---------------------
Catherine Seta et al.
Thinking & Reasoning, forthcoming
Abstract:
Three experiments demonstrated that decisions resulting in considerable amounts of profit, but missed alternative outcomes of greater profits, were rated lower in quality and produced more regret than did decisions that returned lesser (or equal) amounts of profit but either did not miss or missed only slightly better alternatives. These effects were mediated by upward counterfactuals and moderated by participants' orientation to the decision context. That decision evaluations were affected by the availability and magnitude of alternative outcomes rather than the positivity of actual outcomes is counter to the outcome bias effect — a bias in which decisions are rated more positively when they led to more positive outcomes (despite a priori probabilities associated with the decision outcomes). Experiment 3 demonstrated that these effects represent a bias that occurs even when it is clear that the process by which decisions were made followed rational decision processes. This research suggests that when alternative worlds are even better than the desirable outcomes experienced, affect and cognition may be more strongly linked to the magnitude of alternative realities than to obtained outcomes.
---------------------
Asher Koriat, Shiri Adiv & Norbert Schwarz
Personality and Social Psychology Review, forthcoming
Abstract:
Research on group influence has yielded a prototypical majority effect (PME): Majority views are endorsed faster and with greater confidence than minority views, with the difference increasing with majority size. The PME was attributed to conformity pressure enhancing confidence in consensual views and causing inhibition in venturing deviant opinions. Our results, however, indicate that PME for binary choices can arise from the process underlying confidence and latency independent of social influence. PME was demonstrated for tasks and conditions that are stripped of social relevance; it was observed in within-individual analyses in contrasting the individual's more frequent and less frequent responses to the same item, and was found for the predictions of others' responses. A self-consistency model, which assumes that choice and confidence are based on the sampling of representations from a commonly shared pool of representations, yielded a PME for confidence and latency. Behavioral implications of the results are discussed.