Interpreting weight maps in terms of cognitive or clinical neuroscience: nonsense?

Schrouff, Jessica; Mourão-Miranda, Janaı́na

doi:10.1109/prni.2018.8423944

Cited by 14 publications

(10 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The semi-simulated data used in this work have been designed for other work 73 and are described in detail below. The data and code for generating the simulation are available open-source (https://github.com/JessicaSchrouff/Simulated_ECoG).…”

Section: Methodsmentioning

confidence: 99%

Fast temporal dynamics and causal relevance of face processing in the human temporal cortex

et al. 2020

View full text Add to dashboard Cite

We measured the fast temporal dynamics of face processing simultaneously across the human temporal cortex (TC) using intracranial recordings in eight participants. We found sites with selective responses to faces clustered in the ventral TC, which responded increasingly strongly to marine animal, bird, mammal, and human faces. Both face-selective and face-active but non-selective sites showed a posterior to anterior gradient in response time and selectivity. A sparse model focusing on information from the human face-selective sites performed as well as, or better than, anatomically distributed models when discriminating faces from non-faces stimuli. Additionally, we identified the posterior fusiform site (pFUS) as causally the most relevant node for inducing distortion of conscious face processing by direct electrical stimulation. These findings support anatomically discrete but temporally distributed response profiles in the human brain and provide a new common ground for unifying the seemingly contradictory modular and distributed modes of face processing.

show abstract

Section: Methodsmentioning

confidence: 99%

Fast temporal dynamics and causal relevance of face processing in the human temporal cortex

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Note that whilst some neuroimaging features may have a greater contribution than others, all features with non‐zero weights contribute to the model predictions. Although a highly weighted feature is unlikely to be a false positive given adequate signal‐to‐noise ratio (Schrouff & Mourão‐Miranda, 2018), the contribution of any one single feature should be interpreted with caution. For example, although left TPJ activity in the target condition may be of greatest importance in our hallucinations model, this feature alone only contributes approximately 24% towards predictions.…”

Section: Discussionmentioning

confidence: 99%

Multi‐dimensional predictions of psychotic symptoms via machine learning

Taylor

Larsen

Garrido

2020

Human Brain Mapping

View full text Add to dashboard Cite

The diagnostic criteria for schizophrenia comprise a diverse range of heterogeneous symptoms. As a result, individuals each present a distinct set of symptoms despite having the same overall diagnosis. Whilst previous machine learning studies have primarily focused on dichotomous patient-control classification, we predict the severity of each individual symptom on a continuum. We applied machine learning regression within a multi-modal fusion framework to fMRI and behavioural data acquired during an auditory oddball task in 80 schizophrenia patients. Brain activity was highly predictive of some, but not all symptoms, namely hallucinations, avolition, anhedonia and attention. Critically, each of these symptoms was associated with specific functional alterations across different brain regions. We also found that modelling symptoms as an ensemble of subscales was more accurate, specific and informative than models which predict compound scores directly. In principle, this approach is transferrable to any psychiatric condition or multi-dimensional diagnosis.

show abstract

“…Note that whilst some neuroimaging features may have a greater contribution than others, all features with non-zero weights contribute to the model predictions. Although a highly weighted feature is unlikely to be a false positive given adequate signal-to-noise ratio (60), the contribution of any one single feature should be interpreted with caution. For example, although left TPJ activity in the target condition may be of greatest importance in our hallucinations model, this feature alone only contributes approximately 24% toward predictions.…”

Section: Methodological Considerationsmentioning

confidence: 99%

Multi-dimensional predictions of psychotic symptoms via machine learning

Taylor

Larsen

Garrido

2020

Preprint

View full text Add to dashboard Cite

BackgroundThe diagnostic criteria for schizophrenia comprise a diverse range of heterogeneous symptoms. As a result, individuals each present a distinct set of symptoms despite having the same overall diagnosis. MethodsAlthough machine learning techniques are considered a potential gateway to precision psychiatry, prior work has primarily focused on dichotomous patient-control classification.Instead, we predict the severity of each individual symptom on a continuum. We applied machine learning regression within a multi-modal fusion framework to fMRI and behavioural data acquired during an auditory oddball task in 80 schizophrenia patients. ResultsBrain activity was highly predictive of some, but not all symptoms, namely hallucinations, avolition, anhedonia and attention. Each of these symptoms was associated with alterations in specific functional networks encompassing the ventral and dorsal attention networks, the auditory network, amongst other cortical and subcortical regions. We found that an ensemble of subscale models yielded a two-fold increase in accuracy over single models which predict positive and negative compound scores directly. ConclusionsOur results suggest that modelling symptoms as an ensemble of subscales is more accurate, specific, and informative than the compound-based approach. We provide functional brain maps of model contributions identifying the networks of regions which pertain to each individual symptom. This approach is transferrable to any other psychiatric condition and may also contribute to the development of precision psychiatry.

show abstract

Interpreting weight maps in terms of cognitive or clinical neuroscience: nonsense?

Cited by 14 publications

References 12 publications

Fast temporal dynamics and causal relevance of face processing in the human temporal cortex

Fast temporal dynamics and causal relevance of face processing in the human temporal cortex

Multi‐dimensional predictions of psychotic symptoms via machine learning

Multi-dimensional predictions of psychotic symptoms via machine learning

Contact Info

Product

Resources

About