Perceptual decision
How does the brain trade speed for accuracy when deciding?
Open access · cc by · source: Europe PMC
When people are told to hurry, brain signals of evidence build-up reach the same peak as when they are careful, suggesting the brain turns up overall gain rather than simply lowering its decision threshold.
Study at a glance
- Design
- Human experiment — Within-subject random-dot motion task crossing speed vs accuracy instructions (blocked) with easy vs hard coherence; one variant recorded single-pulse TMS motor-evoked potentials, the other 64-channel EEG; race models fitted to pooled reaction times were used to predict the neural signals.
- N
- The provided text omits the Participants section, so the number of people in the TMS and EEG experiments is not stated; models were fitted to data pooled across participants.
- Population
- Adult volunteers in London performing a motion-direction discrimination task
- Outcome
- Reaction times and accuracy; build-up rate and pre-response amplitude of the MEP difference signal and the centroparietal positivity; fit of free- vs forced-excursion model predictions to these neural signals
Structured fields used in claim comparison tables when every cited study has a complete layer.
Key findings
Behaviour showed the expected tradeoff: faster, more error-prone responses under speed instructions and slower responses on hard trials. Both neural signals built up faster on easy than hard trials, confirming they track accumulation, but neither differed in slope or pre-response amplitude between speed and accuracy instructions. The forced-excursion model's predictions matched the neural data better, with an Akaike weight of 0.994 versus 0.006 for the standard free-boundary version in the TMS experiment, and a significantly smaller prediction error in the EEG experiment.
Methodology
Participants judged whether a cloud of moving dots drifted up or down, under blocked instructions to be fast or to be accurate, with easy and hard motion strengths. In one version the researchers used single TMS pulses over motor cortex during the decision to track rising corticospinal excitability; in the other they recorded 64-channel EEG to track the centroparietal positivity, a signal thought to reflect evidence accumulation. They fitted a race model to reaction times, then re-expressed it in a mathematically equivalent 'forced-excursion' form where the boundary stays fixed and the speed-accuracy change is moved onto accumulation rate and noise, and compared which version better predicted the neural signals.
Limitations
The extracted text omits the participant section, so sample sizes cannot be checked here, and models were fitted to trials pooled across people, so conclusions generalise to these participants' trials rather than to the population. The TMS model advantage was significant with a bias-corrected bootstrap interval but not with a simple percentile interval, so the authors urge caution. The two model variants make identical behavioural predictions, so the evidence rests entirely on the noisy neural signals, and the authors do not claim their model beats urgency-signal models, which fitted about as well in exploratory analyses. TMS both measures and perturbs the motor system.
How this study connects
Role on claims
Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.
Not yet placed on a claim. This paper has study layers, but no concept page yet cites it as support, challenge, or qualifier.
Related papers in this topic
Same topic cluster — not a recommendation engine.