N2N Signal & Synthesis · 21 November 2026
What the Average Misses
Two people can give the same answer to a listening test. What happens when each learns the answer was wrong? Four studies explore what a group average can tell us about an individual, and what changes when we measure a different aspect of the same experience.
One Film, Many Viewers
Van De Water and colleagues analysed fMRI recordings from 48 autistic and 90 non-autistic adults watching six film clips. On average, a model using features of the films predicted brain responses about equally well in both groups. A different question produced a different answer. How closely did one viewer’s pattern resemble another’s? Autistic viewers varied more among themselves, especially in how their brains responded to social features.
A group average can conceal that spread. The scans measure how brain activity tracks features of a film. Learning which moments felt important to a particular viewer would also require asking them.
November 2026 · Journal of Neuroscience paper · SciTechDaily, 20 November
After a Wrong Answer
Zuk and colleagues asked adults to judge which of two tones had the higher pitch, then told them whether they were right. Autistic and non-autistic participants achieved similar accuracy. In the autistic group, the average EEG response distinguished negative from positive feedback less strongly. An easy, correct trial also had less influence on the group’s subsequent choices. The researchers saw the pattern again in a follow-up one to two years later.
In this experiment, scoring the answers alone would have missed how each group used recent information. Extending the result to everyday learning would require evidence from other tasks and settings.
26 October · Journal of Neuroscience paper · Neuroscience News, 2 November
When Diagnostic Categories Overlap
Jiao and colleagues approached variation from another direction. They analysed symptoms and brain connectivity in children and identified three profiles that crossed ADHD and anxiety diagnoses. They tested how well the patterns held up in an independent cohort of 224 children. Finding similar patterns in another sample makes them worth investigating. Using a profile to choose treatment for an individual child would require a prospective clinical test.
16 October · Nature Mental Health paper · SciTechDaily, 18 November
Feeling for Someone and Reading Their Emotion
The Psychedelic Research Portal lists the University of Basel group studying how psychedelics affect people. A November paper from that group gives a particularly clear example of why the choice of measure matters.
In a double-blind crossover study, 16 healthy adults completed different sessions in random order, allowing the team to compare each participant with themselves under placebo. Six hours after mescaline, participants scored higher on emotional empathy and lower on cognitive empathy. In everyday terms, feeling more for the person in a picture could coincide with less accuracy in identifying their emotion. Ketanserin, which blocks serotonin 5-HT2A receptors, blocked the observed effects.
The experiment measured an acute response to images. Whether a person subsequently listens more carefully or offers help would need a separate behavioural measure. An average across both empathy scores could conceal the most informative result.
19 November · Klaiber and colleagues, Journal of Psychopharmacology · Portal research-centre directory
N2N Synthesis
The film study asks how much people differ from one another. The pitch study asks how a person’s next choice changes after feedback. The childhood study asks whether similarities between people cross diagnostic boundaries. The mescaline study compares outcomes within the same volunteers and finds that two aspects of empathy move in opposite directions. Combining their results into a single biological explanation would lose the distinction that makes each finding informative.
A question worth testing. If the same viewers watched a film twice, which responses would recur within each person? Asking viewers which scenes mattered to them would give researchers a second measure to compare with the scans. That design could help distinguish a stable individual response from a response that changes with attention, familiarity or the day of testing.
From the archive. The jumping-spider sleep report makes a related point about measurement. Researchers could test responsiveness and sleep rebound. What a spider might experience during sleep remains an open question.

