Positive Mood + Action = Negative Mood + Inaction: Effects of General Action and Inaction Concepts on Decisions and Performance as a Function of Affect

摘要

General action and inaction concepts have been shown to produce broad, goal-mediated effects on cognitive and motor activity irrespective of the type of activity. The current research tested a model in which action and inaction goals interact with the valence of incidental moods to guide behavior. Over four experiments, participants’ moods were manipulated to be positive (happy), neutral, or negative (angry or sad), and then general action, inaction, and neutral concepts were primed. In , action primes increased intellectual performance when participants experienced a positive (happy) or neutral mood, whereas inaction primes increased performance when participants experienced a negative (angry) mood. Including a control-prime condition, and replicated these results measuring the number of general interest articles participants were willing to read and participants’ memory for pictures of celebrities. replicated the results comparing happiness with sadness and suggested that the effect of the prime’s adoption was automatic. Overall, the findings supported an interactive model by which action concepts and positive affect produce the same increases in active behavior as inaction concepts and negative affect.> class="fig iconblock whole_rhythm clearfix" id="F4" co-legend-rid="lgnd_F4"> class="icnblk_cntnt" id="lgnd_F4">> class="sec"> class="kwd-title">Keywords: goals, self-regulation, mood id="idm140717678861872" class="tsec sec headless whole_rhythm">Human behavior is influenced by specific evaluations of the behavior (; ) but also by temporary concepts and moods that are not objectively connected to the behavior (; href="#R4" rid="R4" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215255">Albarracin et al., 2008; href="#R6" rid="R6" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_533293731">Albarracin, Leeper, & Wang, 2009; href="#R19" rid="R19" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215263">Fishbach & Labroo, 2007; href="#R32" rid="R32" class=" bibr popnode">Schwarz & Clore, 1983; href="#R16" rid="R16" class=" bibr popnode">Clore & Schall, 2005). First, general goals of action and inaction have been shown to impact the amount of cognitive and motor output irrespective of the type of behavior (href="#R4" rid="R4" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215248">Albarracin et al., 2008; href="#R28" rid="R28" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215266">Laran, 2010). People primed with general action words (e.g., go) solve intellectual problems, eat, memorize information, and move to a greater extent than people primed with general inaction words (e.g., stop; href="#R4" rid="R4" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215281">Albarracin et al., 2008). Second, moods and temporary affective reactions influence behavior as well (href="#R16" rid="R16" class=" bibr popnode">Clore & Schnall, 2005). For example, suggesting that negative (vs. neutral) moods trigger uncontrolled eating (href="#R25" rid="R25" class=" bibr popnode">Herman & Polivy, 1980), food intake increases in response to anxiety-provoking films (href="#R17" rid="R17" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215276">Cools, Schotte, & McNallyy, 1992; href="#R31" rid="R31" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215257">Schotte, Cools, & McNally, 1990), task failure (href="#R30" rid="R30" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215285">Ruderman, 1985), and pain anticipation (href="#R24" rid="R24" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215283">Herman & Polvy, 1975). Moreover, some negative moods, particularly those low in arousal (e.g., sadness), can signal inadequate progress at a task and increase effort and attention in various contexts (for a review, see href="#R16" rid="R16" class=" bibr popnode">Clore & Schnall, 2005).One interesting and currently unanswered question concerns how general action and inaction prompts influence behavior while people experience positive or negative moods. To begin, each factor may be additively combined as multiple pieces of information are combined in decisions (href="#R8" rid="R8" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215267">Anderson, 1959, href="#R9" rid="R9" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215270">1970, href="#R10" rid="R10" class=" bibr popnode">1974), albeit not necessarily in an intentional fashion. In that case, the effects of action and inaction concepts may remain independent from the effects of mood, with action concepts producing more active behavioral performance (more problem solving, better recognition memory) than inaction concepts, regardless of the mood people experience (see, e.g., href="#R4" rid="R4" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215291">Albarracin et al., 2008).Other effects are also plausible, however. As general action and inaction prompts have been shown to promote general action and inaction goals (href="#R4" rid="R4" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215280">Albarracin et al., 2008), moods may exert interactive, motivational effects on behavior. Past research has demonstrated that neutral behavioral concepts (e.g., doing puzzles) are adopted as goals when the behavior concepts are coactivated with positive affect (href="#R18" rid="R18" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215277">Custers & Aarts, 2005). Correspondingly, neutral behaviors are abandoned when the behavior concepts are coactivated with negative affect (href="#R1" rid="R1" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215252">Aarts et al., 2007, href="#R2" rid="R2" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215295">2008; href="#R18" rid="R18" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215288">Custers & Aarts, 2005). In this past research, an evaluative conditioning paradigm paired goal concepts (e.g., “doing puzzles”) with positive, negative, or neutral affective stimuli (e.g., “sunshine” in the case of positive, “table” in the case of neutral, and “garbage” in the case of negative). In one of these studies, for example, goal-related words became less accessible when paired with negative than neutral affective primes (href="#R1" rid="R1" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215292">Aarts, Custers, & Holland, 2007). In other studies, means consistent with a primed goal were pursued less when the goal concept was paired with negative than neutral concepts (href="#R1" rid="R1" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215271">Aarts, Custers, & Holland, 2007). These effects, which were not driven by arousal or attention and were replicated in a variety of contexts, may occur when negative moods are paired with general action or inaction concepts.This past work on how affective stimuli influence the adoption of goal concepts assumes that goal concepts are conditioned by the simultaneous presentation of affective concepts. Exposure to the word sunshine leads to a temporary affective reaction that is attached to and shapes the evaluation of a behavioral concept such as working on puzzles (href="#R1" rid="R1" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215293">Aarts, Custers, & Holland, 2007). In Aarts and colleagues’s work, however, affect is primed at the trial level, by incidentally pairing the two words. Thus, it is unclear whether moods that are experienced over several minutes or hours can influence the activation of goals in the same fashion. Prior research in the area of affect suggests that conditioning procedures and moods induced before requesting a response both activate positive and negative concepts that can then be attributed to temporally close behavior concepts (Albarracin & Kumkale, 2004; href="#R26" rid="R26" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215265">Jones, Fazio, & Olson, 2009; href="#R36" rid="R36" class=" bibr popnode">Wyer, Clore, & Isbell, 1999), and there are precedents of mood effects on goal activation (href="#R19" rid="R19" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215272">Fishbach & Labroo, 2007). The empirical research on our specific question, however, remains to be conducted.Considering potential interactions between inaction concepts and negative moods suggests some interesting effects. If a positive mood (or perhaps the lack of a negative mood) signals that a behavioral concept is acceptable, priming action during the experience of a positive mood may produce more activity than priming action during the experience of a negative mood. Correspondingly, as mood can signal adoption and rejection of both action and inaction, priming inaction during the experience of a positive mood may produce less activity than priming inaction during the experience of a negative mood. This action/inaction goal adoption is likely to occur with little or no deliberation (href="#R1" rid="R1" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215273">Aarts, Custers, & Holland, 2007; href="#R19" rid="R19" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215289">Fishbach & Labroo, 2007).Our conceptualization implies that the same mechanisms underlying the connections between mood and action goals should apply to the connections between mood and inaction goals. Such connections predict decreased activity in negative than positive mood conditions when action goals are in place, to be contrasted with increased activity in negative than positive mood conditions when inaction goals are in place. In fact, in an experiment manipulating mood, participants with known scores on an eating-restraint scale had an opportunity to eat raisins or popcorn following an induction of anxious, neutral, or happy moods (href="#R17" rid="R17" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215278">Cools, Schotte, & McNally, 1992). Consistent with our predictions, a negative mood produced more eating in participants with high restraint (perhaps similar to an inaction goal) than in participants with low restraint (perhaps similar to an action goal). In contrast, a positive mood produced less eating in participants with high restraint than in participants with low restraint. Although indirectly, these findings support the hypothesis that positive and negative moods can signal adoption or rejection of a behavioral concept, with rejection of an inaction (e.g., not eating) producing an action.Investigating the influences of mood on general goal adoption is important for at least three reasons. First, our predictions complement past research by specifying further mechanisms by which negative affect can facilitate rather than hinder a given behavior. Second, action and inaction concepts have been shown to influence behaviors that are in the focus of attention (a required task following priming) without establishing the situational moderators of this effect (e.g., a current mood). Third, anger has been assumed to increase activity (href="#R14" rid="R14" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215286">Carver & Harmon-Jones, 2009), but we predict conditions under which anger will decrease activity. The present paper reports four experiments. In href="/pmc/articles/PMC3626447/figure/F1/" target="figure" class="fig-table-link figpopup" rid-figpopup="F1" rid-ob="ob-F1" co-legend-rid="lgnd_F1">Experiment 1, participants were first induced to experience either a positive (happy) mood or a negative (angry) mood, were then primed with action or inaction words, and finally solved a series of math and verbal problems after a brief delay. Unlike perceptual-priming effects, prior research has demonstrated that goal-priming effects persist over a brief delay between priming and performance, whereas conceptual priming is not verified after the same delay (href="#R11" rid="R11" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215247">Bargh et al., 2001). Therefore, as we hypothesized that the effects of the primes would be goal-mediated, we expected them to be visible after this delay.Further, if general action and inaction concepts elicit goals to be respectively active and inactive, the primes should interact with the mood manipulation (href="#R18" rid="R18" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215294">Custers & Aarts, 2005; href="#R19" rid="R19" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215251">Fishbach & Labroo, 2007). General action concepts should produce better problem solving than general inaction concepts when participants are induced to experience a positive (happy) mood but worse problem solving when participants are induced to experience a negative (angry) mood. Moreover, as chronic (neutral) moods are generally positive (href="#R34" rid="R34" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_307215264">Storbeck & Clore, 2008) and lack of negative information is perceived as positive (href="#R13" rid="R13" class=" bibr popnode">Borstein, 1989), neutral moods may only differ from negative (angry) mood conditions.href="/pmc/articles/PMC3626447/figure/F2/" target="figure" class="fig-table-link figpopup" rid-figpopup="F2" rid-ob="ob-F2" co-legend-rid="lgnd_F2">Experiment 2 was designed to provide a replication of the effects of general action primes. The replication was planned using a statistically more efficient 2 × 2 design. Participants were first induced to experience either a positive (happy) or negative (angry) mood, were then primed with action or control words, and finally selected how many articles they wanted to read on a topic. We expected that participants would select more articles when primed with action concepts in positive (happy) mood conditions but fewer articles in negative (angry) mood conditions. Additionally, href="/pmc/articles/PMC3626447/figure/F3/" target="figure" class="fig-table-link figpopup" rid-figpopup="F3" rid-ob="ob-F3" co-legend-rid="lgnd_F3">Experiment 3 was designed to provide a similar replication of the effects of general inaction primes. Participants were first induced to experience either a positive (happy) or negative (angry) mood, were then primed with inaction or control words, and then performed a memory task. We hypothesized that participants would show better memory performance when primed with inaction concepts in the negative (angry) mood conditions than the positive (happy) mood conditions.In the first three experiments, participants were induced to experience either a happy mood or an angry mood, which best controls for the processing (e.g., attentional) effects that distinguish the emotional states of happiness and sadness (href="#R12" rid="R12" class=" bibr popnode">Bodenhausen, 1993; see also href="#R7" rid="R7" class=" bibr popnode">Albarracin & Wyer, 2001; href="#R5" rid="R5" class=" bibr popnode tag_hotlink tag_tooltip" id="__tag_533293729">Albarracin & Kumkale, 2003). Like happiness, anger increases reliance on accessible cognitive representations (e.g., stereotypes) because it elicits high arousal. The use of happiness and anger thus permitted comparing the influence of valence without the arousal differences that are naturally present when one compares happiness and sadness. href="/pmc/articles/PMC3626447/figure/F4/" target="figure" class="fig-table-link figpopup" rid-figpopup="F4" rid-ob="ob-F4" co-legend-rid="lgnd_F4">Experiment 4, however, replicated our results comparing happiness with sadness.