Case Report
Are actions better than inactions? Positivity, outcome, and intentionality biases in judgments of action and inaction

https://doi.org/10.1016/j.jesp.2021.104105Get rights and content

Highlights

  • •

    Despite the adaptive value of both action and inaction, people not only evaluate actions more favorably than inactions but also prefer to engage in them as well (action positivity bias)

  • •

    Preferences for action over inaction tend to be driven by biases of outcome positivity (action outcome bias) and intentionality (action intentionality bias), however, assumed outcome positivity is most influential than assumed intentionality

  • •

    An overall preference for action could become detrimental to health, therefore, understanding the magnitude of this bias in everyday life is vital

Abstract

Behavior varies along a continuum of activity, with effortful behaviors characterizing actions and restful states characterizing inactions. Despite the adaptive value of both action and inaction, we propose three biases that, in the absence of other information, increase the probability that people like, and want to pursue, action more than inaction: An action positivity bias, an action outcome bias, and an action intentionality bias. Across four experiments, participants not only evaluated actions more favorably than inactions (Experiment 1–3) but also chose to engage in actions more than inactions (Experiment 4). This action positivity bias was driven by the two interrelated biases of outcome positivity and intentionality (Experiments 1–3), such that actions (versus inactions) were spontaneously thought of as having more positive outcomes and as being more intentional. Moreover, these outcome differences played a stronger role in the action positivity bias than did the intentionality differences (Experiment 3). As balancing action and inaction is important for healthy human functioning, it is important to understand evaluative biases in this domain. All experiments were preregistered, and one involved a nationally representative sample.

Section snippets

Action positivity, action outcome, and action intentionality biases

According to the classic assumptions in decision making, people make behavioral decisions based on their evaluation of the outcomes of a behavior (Ajzen, Fishbein, & Heilbroner, 1980; Ajzen, Fishbein, Lohmann, & Albarracín, 2019; Bar-Eli, Azar, Ritov, Keidar-Levin, & Schein, 2007; Fishbein & Ajzen, 2011; Karsh, Eitam, Mark, & Higgins, 2016; Osgood, 1962). For example, when people are given the option to choose between A and B, and are informed that both will lead to a negative outcome, the

The relation between the action positivity, action outcome, and action intentionality biases

An important goal of our research involved understanding the relation between the proposed action positivity bias, action outcome bias, and action intentionality bias. On the one hand, these biases could all occur in parallel. The Protestant social ethic that permeates Western cultures prescribes work and condemns laziness (Miller, Woehr, & Hudspeth, 2002). Therefore, the greater effort and intent involved in actions could result in people finding actions more desirable and more conducive to

The present research

This research was designed to create a strong experimental paradigm to examine the concepts of action, outcome, and intentionality. Although some past research suggests associations of action with outcome positivity and intentionality judgments, these biases have not been directly demonstrated nor have the underlying processes been investigated. Therefore, the present research examined these biases and their interrelations by carefully manipulating action and inaction, outcomes, and

Experiment 1

Experiment 1 was designed to empirically assess the presence of an action positivity bias, an action outcome bias, and an action intentionality bias, and to begin to understand the relation between the three. We hypothesized that in the absence of information about the outcome of a behavior (a) actions would be judged as more positive (action positivity bias) and more intentional (action intentionality bias) than inactions and (b) actions would be expected to have more positive outcomes and

Preregistration

The design, hypotheses, and analysis plan were all preregistered at the Open Science Framework (https://osf.io/tb6r2/?view_only=52728760eb9d4582a23189f1283c4f94).

Power analysis

This experiment employed a 2 (behavior: flipping a switch, not flipping a switch) x 3 (outcome: positive, negative, unspecified) between-subjects design. As our hypotheses were presented in univariate terms (predicting main effects and interactions for each outcome separately), this power analysis was based on the univariate results of

Manipulation checks

We first performed checks to determine if each of our manipulations had the intended effect. Results supported the conclusion that all manipulations worked as expected.

The effect of the behavior manipulation. An independent samples t-test was conducted to gauge differences in rated action or inaction across the two behavioral conditions. As intended, the behavior describing an action was perceived as more active (M = 5.69, SD = 1.37) than the behavior describing an inaction (M = 3.35, SD

Discussion

The purpose of Experiment 1 was to empirically evaluate the presence of an action positivity bias, an action outcome bias, and an action intentionality bias, and to begin to understand the relation between the three. We found that participants evaluated actions as more positive, and associated actions with more outcome positivity and intentionality, than inactions. This finding is remarkable because participants could have rationalized that they had good reasons to not flip a switch. However,

Experiment 2

Experiment 2 was designed to assess the degree to which differences in the perceived intentionality of action and inaction lead to corresponding evaluative differences between the two. We hypothesized that (a) when presented with behaviors not described as being intentional or unintentional, actions would be judged as more positive and more intentional than inactions. In contrast, (b) when presented with behaviors described as either intentional or unintentional, intentionality information

Preregistration

The design, hypotheses, and analysis plan were all preregistered at the Open Science Framework (https://osf.io/srb7s/?view_only=fdeed2f3cd0c4166ad16603c460cf2ec).

Power analysis

This experiment employed a 2 (behavior: pressing a button, not pressing a button) x 3 (intentionality: high, low, unspecified) between-subjects design. As our hypotheses were presented in univariate terms (predicting main effects and interactions for each outcome separately), this power analysis was based on the univariate results of a

Manipulation checks

We first performed checks to determine if each of our manipulations had the intended effect. Results supported the conclusion that all manipulations worked as expected.

The effect of the behavior manipulation. An independent samples t-test was conducted to gauge differences in rated action or inaction across the two behavioral conditions. As predicted, the behavior describing an action was perceived as more active (M = 4.90, SD = 1.64) than the behavior describing an inaction (M = 3.44, SD

Discussion

The purpose of Experiment 2 was to assess if perceived intentionality differences contribute to evaluative differences between actions and inactions. The results replicated the action positivity and action intentionality biases found in Experiment 1. Moreover, we found that actions were perceived as more positive than inactions because actions are perceived as more intentional than inactions. Therefore, when actions are associated with lower intentionality and inactions are associated with

Experiment 3

Experiments 1 and 2 assessed the relative roles of outcome and intentionality information on the action positivity bias. However, neither experiment obtained evidence about the relative weight of outcome and intentionality information because neither manipulated both factors within the same experiment. Thus, Experiment 1 showed that spontaneous thoughts about outcomes led actions to appear more positive than inactions, but it is possible that intentionality was equally or more important.

Preregistration

The design, hypotheses, and analysis plan were all preregistered at the Open Science Framework (https://osf.io/gpvue/?view_only=235f2e8ef35d42018de4ce7a4554ed72).

Power analysis

This experiment employed a 2 (behavior: flipping a switch, not flipping a switch) x 3 (outcome: positive, negative, unspecified) x 3 (intentionality: high, low, unspecified) between-subjects design. As our hypotheses were presented in univariate terms (predicting main effects and interactions for each outcome separately), this power

Manipulation checks on ratings of action/inaction and evaluations

We first performed checks to determine if each of our manipulations had the intended effect. Results supported the conclusion that all manipulations worked as expected.

The effect of the behavior manipulation. An independent samples t-test was conducted to gauge differences in rated action or inaction across the two behavioral conditions. As expected, the behavior describing an action was perceived as more active (M = 5.01, SD = 1.85) than the behavior describing an inaction (M = 4.08, SD

Discussion

The purpose of Experiment 3 was to assess the role of both outcome and intentionality in evaluations of action and inaction. The findings showed that participants evaluated actions as more positive and more intentional than inactions, replicating prior experiments. However, when outcome and intentionality information both varied, the effect of outcome information on evaluations was stronger, implying that the action outcome bias dominates over the action intentionality bias.

Experiment 4

Experiment 4 was designed to test the implications of the action positivity bias on behavior, by examining whether evaluations favoring action would similarly translate into preferences for active behaviors. We hypothesized that, (a) when given the opportunity to engage in an action or an inaction, actions would be preferred, suggesting that biases for action extend to behaviors as well. Consistent with prior experiments, we further hypothesized that, (b) when asked to evaluate the behavior

Preregistration

The design, hypotheses, and analysis plan were all preregistered at the Open Science Framework (https://osf.io/7pu5z/?view_only=9a8e4a5df403488d8bd90cd7eca8cb75).

Power analysis

This experiment employed a 2 (behavior: pressing a button, not pressing a button) between-subjects design. The power analysis was based on results of a pilot study that employed the same design. A df = 0.29 (a small-to-medium effect, according to Cohen's, 1992 effect size convention) was chosen because it was the size of the effect

Manipulation checks on ratings of action/inaction and effort

An independent samples t-test was conducted to examine differences in rated action or inaction across the two experimental conditions. As intended, the task involving an action was perceived as more active (M = 2.95, SD = 1.04) than the task involving an inaction (M = 1.50, SD = 0.93), t(139) = 6.32, p < .001, d = 1.47. The same held true for effort; for action: M = 2.19, SD = 1.17; for inaction: M = 1.58, SD = 1.02; t(139) = 2.35, p = .02, d = 0.56.

Behavior preference

A chi-square goodness of fit test was

Discussion

The purpose Experiment 4 was to investigate whether conclusions about the action positivity bias had implications for behavioral preferences. Experiment 4 extended the results from previous experiments, showing that people choose action over inaction. Once people had engaged in the task, however, the actual experience with the task (i.e., pressing or not pressing a button) was evaluatively neutral. As actual experience with a behavior should shape evaluations based on the nature of that

Author note

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors certify that they have no competing interests that could influence their work. Correspondence concerning this article should be addressed to Aashna Sunderrajan, Department of Psychology, University of Illinois at Urbana-Champaign, IL 61820. Contact: [email protected].

Open practices

All experiments in this paper were preregistered. For the preregistration plans, see: https://osf.io/tb6r2/?view_only=52728760eb9d4582a23189f1283c4f94 (Experiment 1), https://osf.io/srb7s/?view_only=fdeed2f3cd0c4166ad16603c460cf2ec (Experiment 2), https://osf.io/gpvue/?view_only=235f2e8ef35d42018de4ce7a4554ed72 (Experiment 3), and https://osf.io/7pu5z/?view_only=9a8e4a5df403488d8bd90cd7eca8cb75 (Experiment 4).

References (58)

  • I. Ajzen et al.

    The influence of attitudes on behavior

  • D. Albarracín

    Action and inaction in a social world: Prediction and change of attitudes and behavior

    (2020)
  • D. Albarracín et al.

    A test of major assumptions about behavior change: A comprehensive look at the effects of passive and active HIV-prevention interventions since the beginning of the epidemic

    Psychological Bulletin

    (2005)
  • D. Albarracín et al.

    Action, inaction, and actionability: Definitions and implications for communications and interventions to change behaviors

  • D. Albarracín et al.

    Immediate increase in food intake following exercise messages

    Obesity

    (2009)
  • D. Albarracín et al.

    Action dominance: The performance effects of multiple action demands and the benefits of an inaction focus

    Personality and Social Psychology Bulletin

    (2018)
  • E. Aronson et al.

    The effect of severity of initiation on liking for a group

    The Journal of Abnormal and Social Psychology

    (1959)
  • J. Baron et al.

    Outcome bias in decision evaluation

    Journal of Personality and Social Psychology

    (1988)
  • R.G. Benoit et al.

    Forming attitudes via neural activity supporting affective episodic simulations

    Nature Communications

    (2019)
  • J.W. Brehm

    A theory of psychological reactance

    (1966)
  • R. Brown et al.

    The psychological causality implicit in language

    Cognition

    (1983)
  • J. Cohen

    Things I have learned so far

    American Psychologist

    (1990)
  • J. Cohen

    A power primer

    Psychological Bulletin

    (1992)
  • C. Darviri et al.

    The healthy lifestyle and personal control questionnaire (HLPCQ): A novel tool for assessing self-empowerment through a constellation of daily activities

    BMC Public Health

    (2014)
  • J.R. Detert et al.

    Moral disengagement in ethical decision making: A study of antecedents and outcomes

    Journal of Applied Psychology

    (2008)
  • A. Fagerlin et al.

    Cure me even if it kills me: Preferences for invasive cancer treatment

    Medical Decision Making

    (2005)
  • R.H. Fazio et al.

    Direct experience and attitude-behavior consistency: An information processing analysis

    Personality and Social Psychology Bulletin

    (1978)
  • A. Fillon et al.

    Impact of past behavior normality: Meta-analysis of exceptionality effect

    Cognition and Emotion

    (2020)
  • M. Fishbein et al.

    Predicting and changing behavior: The reasoned action approach

    (2011)

    • Cited by (8)

    • How job crafters are selected during recruitment: An invisible filter based on job crafting experiences and applicant gender

      2024, Journal of Business Research

    • Is there a difference between stopping and avoiding? A review of the mechanisms underlying Go/No-Go and Approach-Avoidance training for food choice

      2023, Current Opinion in Behavioral Sciences

    • Climate Games: Experiments on How People Prevent Disaster

      2024, Climate Games: Experiments on How People Prevent Disaster

    • Default Matters in Trust and Reciprocity

      2023, Games

    • How Social Media Algorithms Shape Offline Civic Participation: A Framework of Social-Psychological Processes

      2023, Perspectives on Psychological Science

    • Communication Climate at Work: Fostering Friendly Friction in Organisations

      2023, Communication Climate at Work: Fostering Friendly Friction in Organisations
    View all citing articles on Scopus
    View full text
    © 2021 Elsevier Inc. All rights reserved.