# SI-Neural-Computations: fMRI of Social Interaction Recognition with PHASE Animations

## Facts

- Provider: Manasi Malik, Minjae Kim, Tianmin Shu, Shari Liu, Leyla Isik
- License: CC0-1.0 (https://openneuro.org/datasets/ds008495)
- Stored at: Amazon S3, s3://openneuro.org/ds008495 (us-east-1)
- Access: public, readable without an account
- DOI: https://doi.org/10.18112/openneuro.ds008495.v1.0.2
- Category: Neuroscience
- Source: https://doi.org/10.64898/2026.02.20.707082
- Page: https://scigantic.com/archive/1a43594a-f01e-5d37-9e55-be07e39a1ab0
- Keywords: Neuroscience, fMRI, AWS Open Data

## About

fMRI and behaviour from 29 adults (ages 18 to 35; 25 analysed after preregistered exclusions) accompanying Malik, Kim, Shu, Liu and Isik (2026), Bottom-up and generative computations uniquely explain neural responses across the social brain. In the main task (task-main) participants watched 50 ten-second animated videos from the PHASE dataset of procedurally generated interactions between two agents (friendly, neutral and adversarial), mostly over 10 runs. Localizers: social interaction perception with point-light figures (task-sipsts, STS), a false-belief theory of mind localizer (task-tom) and a physics localizer (task-physics). Post-scan relationship judgments are in beh/, and participants.tsv records inclusion and exclusion reasons. BIDS layout, 1,486 files, about 16 GB; preregistration on OSF (osf.io/hq3r7). Useful for comparing vision models and inverse-planning models against social brain responses in the superior temporal sulcus and TPJ. Released under CC0 on OpenNeuro; cite Malik et al. 2026 and the dataset DOI.

## Load the data

```python
import boto3
from botocore import UNSIGNED
from botocore.client import Config
s3 = boto3.client('s3', region_name='us-east-1', config=Config(signature_version=UNSIGNED))
s3.list_objects_v2(Bucket='openneuro.org', Prefix='ds008495/', Delimiter='/')
```

Lists the files with anonymous access. Runs on any machine, with no cloud account.

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/1a43594a-f01e-5d37-9e55-be07e39a1ab0
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/1a43594a-f01e-5d37-9e55-be07e39a1ab0
- Over MCP: `get_schema_card` with id `1a43594a-f01e-5d37-9e55-be07e39a1ab0` at https://api.scigantic.com/mcp

## Related topics

- [Neuroscience datasets](https://scigantic.com/topics/neuroscience)
- [fMRI datasets](https://scigantic.com/topics/fmri)
