Error management, the speaker-bias, and lie-rarchy in the evolution of deception avoidance

Larson, Michael (2012) Error management, the speaker-bias, and lie-rarchy in the evolution of deception avoidance. PhD thesis, James Cook University.

[img]
Preview
PDF (Thesis)
Download (1MB) | Preview
 
73


Abstract

Decades of research have yet to provide a comprehensive theory explaining how, without adequate detection skills, humans circumvent costly deception. This series of studies proposed that biased perceptions of deception evolved to avoid costly deception without having to accurately detect it, per se. Based on evolutionary and error management models, three research studies were conducted to test the proposal that cost-benefit analyses and mental shortcuts produce biased perceptions of deception. In particular, the studies explored the role of the receiver of a message, characteristics of the speaker, and the cost associated with the message in the over- or under-perception of deception.

Study 1 explored if the receiver's characteristics (sex and relationship status), the content of the message, and priming mate value influenced perceptions of veracity. A sample of university students and individuals from the wider community (N = 104) completed a word-priming task where they reordered a series of short scrambled sentences containing words related to high or low mate value. Following priming, participants judged the veracity of messages considered to carry higher or lower reproductive costs. The findings indicated that if costs were minimal individuals maximised the frequency of Type II errors (false-negative) resulting in greater perceptions of honesty. However, the process typically producing truth-biased perceptions began to generate amplified perceptions of deception as costs increased. These findings are consistent with an error management approach, indicating that individuals maximised the number of Type I errors (false-positive) as costs increased.

Study 2 explored perceptual biases that may arise based on the characteristics of the speaker. This study proposed a speaker-bias in deception detection; specifically, that initial perceptions of truthfulness vary as a function of a speaker's sex and attractiveness. Participants (N = 117) viewed a series of written vignettes involving an anonymous 'speaker' revealing biographical details to a 'receiver' character. After each vignette, participants were informed as to the truthfulness of the message (either truthful or a lie) and asked to indicate the anonymous speaker character from a series of photos of individuals varying in sex and facial attractiveness. As expected, participants rated female speakers as a more likely source of truthful statements and male speakers as a more likely source of potential deception. It was also found that photos of attractive individuals were selected more frequently during honest conditions.

Study 3 resulted in the construction of a hierarchy of lies based on participants' preferences to avoid different types of deception. This study also allowed participants to make choices among various forms of deception simultaneously to directly investigate the fundamental trade-offs between preferences to avoid different types of deception. Each participant was given a list of ten traits an individual might lie about to secure a new partner. Participants (N = 147) were provided with a high (30 tokens) and low (10 tokens) 'budget'. Participants allocated as many or as few of the tokens as they wished across the traits to indicate their preference not to be deceived regarding each trait. Based on theoretical notions, the ten traits were categorised into one of four tiers a priori. As expected, traits categorised into the higher tiers produced higher avoidance preferences, while traits in lower traits produced lower avoidance preferences. Further, the preference to avoid higher tier traits was most evident when greater restrictions were placed on the participant's selection process. However, as budget restrictions were lifted, trade-offs were found to occur, indicating an increased distribution of resources to lower tiers. Taken together, the findings from the three studies conducted provide a foundation for a new cognitive model of deception avoidance that may have greater explanatory power than earlier models.

Item ID: 40014
Item Type: Thesis (PhD)
Keywords: cognitive science; deception; detection; honesty; lies; mate selection; perception; psychology; truth
Date Deposited: 12 Aug 2015 22:37
FoR Codes: 17 PSYCHOLOGY AND COGNITIVE SCIENCES > 1701 Psychology > 170109 Personality, Abilities and Assessment @ 33%
17 PSYCHOLOGY AND COGNITIVE SCIENCES > 1701 Psychology > 170110 Psychological Methodology, Design and Analysis @ 33%
17 PSYCHOLOGY AND COGNITIVE SCIENCES > 1701 Psychology > 170112 Sensory Processes, Perception and Performance @ 34%
SEO Codes: 97 EXPANDING KNOWLEDGE > 970117 Expanding Knowledge in Psychology and Cognitive Sciences @ 100%
Downloads: Total: 73
Last 12 Months: 5
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page