How fast does the brain combine intentions and outcomes in moral judgment?
People weigh intentions heavily when judging harm, and brain responses separate successful from attempted acts in stages within the first second.
Source
Temporal Dynamics of the Integration of Intention and Outcome in Harmful and Helpful Moral Judgment
Study at a glance
- Design
- Human experiment — Two within-subject EEG/ERP experiments crossing intention (valenced vs neutral) with outcome (valenced vs neutral) in harm and help stories
- N
- N=55 · 55 students total: 27 in Experiment 1 (harm) and 28 in Experiment 2 (help)
- Population
- Right-handed undergraduates at Beijing Normal University
- Outcome
- Permissibility (harm) or praise (help) ratings, reaction times, and ERP amplitudes over frontal and temporo-parietal sites
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
Students read short stories in which an agent intended to cause harm (or help) or not, and did or did not produce that outcome, giving four conditions: none, accidental, attempted and successful. One group judged how permissible harmful acts were; a separate group judged how much praise helpful acts deserved. EEG was recorded time-locked to the key word revealing the agent's intention.
What they found
For harm, negative intentions and negative outcomes both made acts less permissible, and they interacted: accidental harm was judged much worse than no harm, more than successful was compared with attempted harm. For help, intention and outcome each raised praise but did not interact. In both experiments an early component around 180 ms over right temporo-parietal sites was larger for successful than attempted acts, a component near 250 ms over the left side was larger for attempted acts, and later frontal and temporo-parietal waves also separated attempted from successful acts.
The limits
What it doesn't show
The samples were small (under 30 per experiment) and the harm and help experiments used different people and different questions (permissibility versus praise), so the harm-help asymmetry could reflect the task rather than valence. ERPs have poor spatial resolution, so linking the effects to the temporo-parietal junction is an inference, not a measurement. The authors note that some harm stories hinted at the agent's intention before the keyword appeared, and that encoding intention and integrating it with outcome were not cleanly separated in time.
Key terms
- Event-related potential (ERP)
- Average EEG voltage change time-locked to a specific event, giving millisecond timing of brain processing.
- Attempted harm
- The agent meant to cause harm but the outcome was neutral.
- Accidental harm
- The agent meant no harm but a harmful outcome occurred.
- Temporo-parietal junction (TPJ)
- A brain region where temporal and parietal lobes meet, linked to thinking about other people's mental states.
- Intention-outcome integration
- Combining what someone meant to do with what actually happened to reach a moral judgment.
Flashcards
0 of 10 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
In the harm experiment, what happened to permissibility ratings?
Common questions
Why record EEG rather than fMRI here?
EEG has millisecond timing, so it can show when intention and outcome are combined, whereas earlier fMRI work showed only where.
Is judging help simply the mirror image of judging harm?
Not in this study: for harm, intention and outcome interacted, but for help they added up independently, although different questions were asked in each experiment.
What three stages do the authors propose?
A fast right temporo-parietal response to morally extreme information, then an early left temporo-parietal representation of intention, and finally later frontal and temporo-parietal integration and reasoning.
More on Moral psychology