attck-research

Evidence review: Mood and Decision-Making

Article: Mood and Decision-Making
Reviewed: November 2026 · Method version 2 · How these reviews work

How this review works

Each factual claim in the article was checked against the source it cites, then challenged by reviewers whose job was to find the strongest evidence against it: replication attempts, meta-analyses, later papers that cite the original work, and research from other fields that study the same mechanism. A separate reviewer weighed both sides, and a script re-checked every quoted source. Where a claim was overstated or misattributed, the article was corrected. The research below supports the article as it now reads.

Background research

Introduction

A 2015 review in the Annual Review of Psychology concluded that emotions are potent, pervasive and predictable drivers of decisions, sometimes harmful and sometimes helpful.

Lerner and colleagues’ 2015 review is the standard summary of this literature; its conclusion is reported accurately.

The Mood Users Bring With Them

Moods carried in from unrelated events (a bad commute, a long meeting) follow the user into unrelated decisions such as a signup form.

Schwarz and Clore note that moods are influential precisely because people rarely notice where they came from; adding the marker to this paragraph would credit that.

Finding 1

A passing ‘By the way, how’s the weather down there?’ was enough to stop the rainy-day gloom from coloring the answers.

The discounting effect appeared only for sad moods in both of the 1983 experiments, and rests on small samples.

Once the real source of a bad mood is obvious, it stops counting as evidence.

In Schwarz and Clore’s studies, pointing out the cause of a mood removed its influence only for bad moods.

A Security Alert

Lerner and Keltner found that fearful people gave pessimistic risk estimates and made risk-averse choices, while angry people gave optimistic estimates and took risks.

A 2020 meta-analysis (Wake, Wormwood and Satpute) confirms that fear lowers risk taking and raises risk estimates, with a small-to-moderate effect.

The angry user in the same flow

Now the alert interrupts someone halfway through a purchase. Angry people tend to feel certain and in control, and Lerner and Keltner’s results suggest an optimistic read of the risk: tap “This was me” and move on.

Ferrer and colleagues (2017) found incidental anger drives risk taking more in men than in women.

What changed in the article

These edits were made on November 12, 2026 after the review.

Finding 1: contested

Claim reviewed: People were more satisfied with their lives on sunny days, but only when their attention had not been drawn to the weather.

Now reads: People were more satisfied with their lives on sunny days, but only when their attention had not been drawn to the weather. It was a small study, and later work with far more people found the weather effect much smaller or missing, so read it as a vivid demonstration of the mechanism rather than a measure of its size.

Why: The article reports the 1983 phone study accurately, and the mechanism it illustrates (people read a passing mood as an answer unless its source is pointed out) is well established. But the weather finding itself has not held up in larger tests: nine replications with bigger samples were mostly null, a sample of over a million Americans found daily weather mostly unrelated to life satisfaction, and diary studies find that weather’s average effect on mood itself is small, with large individual differences. One longitudinal Australian panel does find a robust day-to-day effect, so the evidence points both ways while the article presents the finding as settled. Now cites Yap et al., The effect of mood on judgments of subjective well-being: Nine tests of the judgment model (JPSP 2017).

Finding 1: supported with caveats

Claim reviewed: When people evaluate something they implicitly ask how they feel about it, and can mistake whatever they happen to feel for a reaction to the thing being judged.

Now reads: Twenty years later, the authors looked back on the mechanism.

Why: The mechanism is described faithfully, and it is the central idea of the feelings-as-information account. The only slip is timing: the authors proposed it in 1983 and restated it twenty years later, so ‘Twenty years later, the authors described the mechanism’ slightly misplaces where it came from.

Finding 1: supported with caveats

Claim reviewed: Once the real source of the mood is obvious, it stops counting as evidence.

Now reads: Once the real source of a bad mood is obvious, it stops counting as evidence.

Why: The principle is in the source, but in both of the original experiments the discounting worked only for bad moods; good moods kept coloring judgments even when their cause was pointed out. The sentence states it for every mood.

Finding 2: contested

Claim reviewed: Cheerful users skim and are easy to satisfy; the people who read every word are often having a bad day, so weak reasoning survives the first audience and fails the second.

Now reads: That sets up an awkward asymmetry. In Bless’s study, cheerful people skimmed, though later work found they read closely when the message was pleasant or when the bad news concerned them directly. The people most likely to read every word are often having a bad day: the customer disputing a charge, the account holder who just got a warning. Weak reasoning can survive a cheerful reader and fail an unhappy one.

Why: The sentence turns one 1990 lab result into a rule about users, and the wider literature does not support a rule. Happy people did skim in several studies, including when they were told to stop reading once they had enough information, but others found them reading closely when the message was pleasant or when the bad news was about their own habits, a meta-analysis favored that conditional account, and the current theoretical view is that the link between mood and processing depth is malleable and often reversed. ‘Easy to satisfy’ was never tested; the sturdier half of the claim is that people in a bad mood scrutinize reasons. Now cites Wegener, Petty & Smith, Positive mood can increase or decrease message scrutiny (JPSP 1995).

A Security Alert: supported with caveats

Claim reviewed: Feelings of certainty and control explained much of the difference between fear and anger.

Now reads: Feelings of certainty and control helped explain the difference, with control doing most of the work.

Why: The paper found that certainty and control moderated the effects, but only control was shown to carry them. ‘Explained much of the difference’ is a little stronger than the abstract.

The choice / recovery steps: misattributed

Claim reviewed: A frightened, risk-averse reader leans toward the safe answer even when the sign-in was theirs, and (per Bless) reads ‘for your security’ closely.

Now reads: Control comes back through small, visible steps, one at a time with progress shown. Fear is an uncertain state, and people who feel uncertain tend to read carefully rather than take a message on trust, so “for your security” will be read closely and found empty. Each step needs a specific reason.

Why: The first half holds: frightened people lean toward the safe choice, and a meta-analysis of 68 samples confirms it. The second half credits Bless, who studied sad moods, not fear; the close-reading prediction is instead supported by research on emotional uncertainty, which finds that emotions marked by uncertainty, fear among them, push people toward careful processing and attention to argument quality, and that inducing fear about a risk led people to process a message about it systematically. The evidence on fear is not unanimous (one experiment found anger produced deeper processing than fear, and anxiety research finds attention pulled toward the threat), so the sentence should cite the uncertainty research and drop the Bless attribution. Now cites Tiedens & Linton, Judgment under emotional certainty and uncertainty: The effects of specific emotions on information processing (JPSP 2001).

The angry user in the same flow: supported with caveats

Claim reviewed: An angry user interrupted mid-purchase feels certain and in control, and Lerner and Keltner’s results predict an optimistic read of the risk (‘This was me’).

Now reads: Now the alert interrupts someone halfway through a purchase. Angry people tend to feel certain and in control, and Lerner and Keltner’s results suggest an optimistic read of the risk: tap “This was me” and move on.

Why: The article frames this as a prediction, which is fair: lab and field studies both found angry people see less risk. But ‘feels certain and in control’ is a tendency, not a measured state of every angry user, and the anger effect is markedly stronger in men than in women.

Conclusion: supported with caveats

Claim reviewed: Mood decides how generous people are with a product, how closely they read and how much risk they accept.

Now reads: No one arrives at a screen in a neutral state. Mood shapes how generous people are with your product, how closely they read and how much risk they accept.

Why: Each part of the sentence rests on research the article cites, but ‘decides’ overstates it. The effects are real and predictable in direction, yet usually small to moderate, and some replications found little effect at all.