Contribution
Two lines of work disagreed about positive affect in goal pursuit: as a “go” signal that boosts commitment (Clore et al. 2001; Custers & Aarts 2005; Fishbach & Labroo 2007) and as a sign of sufficient progress that lets the goal be deactivated (Carver & Scheier 1990, 1998; Carver 2003). This paper proposes and shows a moderator: competing goals. Positive affect lowered a goal’s activation when others were salient and raised it when they were not.
Key insight
Positive affect does not carry a fixed instruction about the goal at hand. It is read in light of what else needs resources: with rivals waiting, as “enough progress, move on”; with none, as “this is going well, keep going”. The authors call this motivated use of affective information.
Assumptions
- Goals are knowledge structures whose activation varies and is deactivated or inhibited once a goal is attained or enough progress has been made (Introduction).
- Activation is read from lexical-decision speed to a word standing for the goal (Shah & Kruglanski 2003), faster meaning more active (Experiment 1, Goal activation), and in Experiment 2 from two intention items.
- Affect as information, interpreted by motivated reasoning (Kunda 1990; Kruglanski): “the presence of competing goals will motivate the actor to use positive affect as information indicating that sufficient progress towards the focal goal has been made” (Competing goals).
- The Carver side is stated from Carver and Scheier and Carver 2003 (Positive affect): “when such a threshold is reached, the actor deactivates the goal. This model suggests that when positive affect is experienced, it should lead to coasting on the focal goal in the form of diminished effort (Carver 2003).”
Key results
- Experiment 1 (n = 76 University of Maryland undergraduates). Each thought for 2 minutes about something to do for someone close, typed one word for it, and wrote a promise beginning “I promise…”. Competing goals: list three things “would like to accomplish today”, against three things “had already accomplished today”. Affect: five IAPS pictures, 12 s each, positive (e.g. a smiling baby) or neutral (e.g. a blue towel). Then a lexical decision task with the goal word presented once among five neutral words and six non-words; log latencies, neutral-word latency as covariate.
- Experiment 1, results. Competing goals × affect, F(1,71) = 7.50, p < .01. With competing goals, the goal word was slower after positive pictures (680 ms) than after neutral ones (595 ms), t(71) = 2.11, p < .05; without them, faster (561 against 627 ms), t(71) = 2.02, p < .05. Competing goals mattered after positive affect (680 against 561 ms, t(71) = 2.82, printed as p < .001) and not after neutral, t(71) = 1.24, ns.
- Experiment 2 (n = 78). Same, but the control listed three colours, and the outcome was two intention items, 1–9, r = .94. Main effect of competing goals, F(1,74) = 5.58, p < .05 (7.73 against 8.32). Interaction F(1,74) = 10.23, p < .01. With competing goals, intentions were weaker after positive affect (7.30 against 8.16), t(74) = 2.39, p < .05; without them, stronger (8.70 against 7.95), t(74) = 2.10, p < .05. Competing goals mattered after positive affect, t(74) = 3.95, p < .001, and not after neutral, t(74) = .58, ns.
- Manipulation check (Experiment 1, Affect manipulation). In a separate sample of 50, the positive set gave a more positive mood (5.12) than the neutral set (4.52) on a 7-point scale, t(48) = 2.04, p < .05. Mood was not measured within either experiment.
- Conclusion (General discussion). “When the actor is motivated to infer progress because of the competition for resources from alternative goals, positive affect is more likely to lead to decreased goal activation. However, when the actor is motivated to infer commitment because the focal goal is the sole concern, positive affect is more likely to lead to increased goal activation.”
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Positive affect lowers a promised goal’s activation when competing goals are salient | moderate | Exp. 1 RT and Exp. 2 intentions, simple effects p < .05 |
| C2 | Positive affect raises it when no competing goals are salient | moderate | same, opposite direction, p < .05 |
| C3 | Without positive affect, competing goals do not change the goal’s activation | weak | null simple effects (Exp. 1 t = 1.24; Exp. 2 t = .58) |
| C4 | Competing goals moderate positive affect by motivating an inference of progress rather than of commitment | weak | the inference is posited, not measured |
| C5 | The pattern would hold for behaviour | asserted | argued from goal-priming work; behaviour not measured |
Concepts
- Goal activation — the momentary accessibility of a goal’s representation, here the speed of recognising its word.
- Goal deactivation — inhibition of a goal’s representation once enough progress is inferred, freeing resources for other goals (Kruglanski et al. 2002).
- Competing goals — other goals made salient by listing what one wants to accomplish later that day.
Connections
It sets the Carver line (coasting; cited for Carver 2003, LIT-1445, which the record has not read) against the affect-as-information line (Clore et al. 2001; Fishbach & Labroo 2007), and reconciles them with Kruglanski’s goal-systems theory. Its moderator is the same one Thürmer, Scheier and Carver (LIT-1878, read the same day) take from Carver 2003 and Gollwitzer and Rohloff: coasting needs another goal to shift to.
Bearing on the record
- CLAIM-468: partly, and qualifies it. The objection’s positive half says positive affect “is reported to lower a goal’s priority”. This paper finds that it does only when competing goals are salient, and that with none it raises the goal’s activation and the intention to pursue it. So the positive half holds conditionally. The finding cuts both ways for the objection: it supports its conclusion that the relation between valence and the ordering’s hold “can run either way”, and it says the condition the clause would need is competition among goals. Two limits: the affect was incidental (pictures, not progress on the goal), so it is not goal-conduciveness valence about that goal; and activation is a weaker proxy for priority than effort or performance.
- CLAIM-440: partly. Its clause 3 waits on whether positive affect lowers a goal’s priority. Here the lowering appears exactly when goals compete, the condition under which the claim’s account defines valence at all (contested pairs, TERM-076’s nonvacuity). Whether a lowered focal goal is a weaker hold of the prevailing ordering or a move that ordering makes is the claim’s question; the paper does not decide it.
- No instruction for machine-learning practice; nothing for the anthology.
Limitations
- One trial per person. In Experiment 1 the goal word was presented once, so each person’s activation score is a single reaction time.
- Incidental affect. Pictures, not goal progress, produced the affect; the paper is about affect as information, not about the affect a goal’s own progress generates.
- Mood not checked in the experiments; the pilot difference was modest (5.12 against 4.52).
- Different controls. Experiment 1’s control listed goals already completed today, which may itself signal progress; Experiment 2’s listed colours.
- Reporting slips. The dispersion figures in Experiment 2 are labelled SD (.18, .25, .26) but are of the size of standard errors; t(71) = 2.82 is printed as p < .001, where it is about p = .006.
- Behaviour not measured (General discussion).
Open questions
- Does positive affect generated by progress on the goal itself show the same moderation? That is the case Carver’s account concerns.
- Is the moderator competition as such, or the salience of unfinished goals? A design that varied how many goals compete would separate them.