Common population codes produce extremely nonlinear neural manifolds

De, Anandita; Chaudhuri, Rishidev

doi:10.1073/pnas.2305853120

Cited by 5 publications

(5 citation statements)

References 93 publications

(171 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…What happens if these assumptions are violated? Violations of the first and second assumptions—independence and linearity of patterns—are known to impact data analysis ( 23 ) and are discussed at length in statistics textbooks ( 24 ). Namely, when the latent underlying patterns are not independent of each other, PCA will output components which fail to capture any of the correlated factors ( Fig.…”

Section: Resultsmentioning

confidence: 99%

“…It is also unknown how phantom oscillations impact other machine learning methods that use PCA for preprocessing. Many studies explore how low-level statistics influence dimensionality reduction ( 21 – 23 , 32 , 50 – 58 ). These studies mirror our own by showing that patterns in high-dimensional data analysis may not always reflect true patterns in the data.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Phantom oscillations in principal component analysis

Shinn

2023

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Principal component analysis (PCA) is a dimensionality reduction method that is known for being simple and easy to interpret. Principal components are often interpreted as low-dimensional patterns in high-dimensional space. However, this simple interpretation fails for timeseries, spatial maps, and other continuous data. In these cases, nonoscillatory data may have oscillatory principal components. Here, we show that two common properties of data cause oscillatory principal components: smoothness and shifts in time or space. These two properties implicate almost all neuroscience data. We show how the oscillations produced by PCA, which we call “phantom oscillations,” impact data analysis. We also show that traditional cross-validation does not detect phantom oscillations, so we suggest procedures that do. Our findings are supported by a collection of mathematical proofs. Collectively, our work demonstrates that patterns which emerge from high-dimensional data analysis may not faithfully represent the underlying data.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Phantom oscillations in principal component analysis

Shinn

2023

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

show abstract

“…Clearly, MP neurons’ width, indexed by an average of the FWHM across all peaks, was significantly narrower than SP neurons’ width (RNN: 1.29 versus 2.14, p < 0.001; PoS: 0.82 versus 1.05, p < 0.001; ADn: 0.79 versus 1.28, p < 0.001). Recent studies have suggested that the width of a neuron’s tuning curve plays a critical role in shaping the dimensionality of neural representation spaces (De and Chaudhuri, 2023; Kim et al, 2020; Kriegeskorte and Wei, 2021; Langdon et al, 2023), which in turn influences the functionality of neural networks. Thus, here we delved into the impact of tuning curve’s width on HD and AHV encoding.…”

Section: Resultsmentioning

confidence: 99%

“…The key factor for expanding the dimensionality of neural representational spaces is the width of tuning curves. Research has repeatedly shown that, for one-dimensional features, narrower tuning curves not only enable individual neurons to encode a greater volume of information (Zhang and Sejnowski, 1999) but also contribute to increasing the linear dimensions of neural populations (De and Chaudhuri, 2023; Kriegeskorte and Wei, 2021) and producing a more reliable signals for low latency communication (Lenninger et al, 2023). Note that this characteristic has previously been discussed primarily in the context of SP neurons, which exhibit a great variance in their tuning curve widths.…”

Section: Discussionmentioning

confidence: 99%

Encoding of interdependent features of head direction and angular head velocity in navigation

Cai,

Liu,

Liu

2024

Preprint

View full text Add to dashboard Cite

To comprehend the complex world around us, our brains are tasked with the remarkable job of integrating multiple features into a cohesive whole. While previous studies have primarily focused on the processing and integration of independent features, here we investigated the simultaneous encoding of the interdependent features, specifically head direction (HD) and its temporal derivative, angular head velocity (AHV), by first employing computational modeling on HD systems to explore emergent algorithms and then validated its biological plausibility with empirical data from mice's HD systems. Our analysis revealed two distinct neuron populations: those with multiphasic tuning curves for HD compromised their HD encoding capacity to better capture AHV dynamics, while those with monophasic tuning curves primarily encoded HD. This pattern of functional dissociation was observed in both artificial HD systems and the cortical and subcortical regions upstream of biological HD systems, suggesting a general principle for encoding interdependent features. Further, exploration of the underlying mechanisms involved examining neural manifolds embedded within the representational space constructed by these neurons. We found that the manifold by neurons with multiphasic tuning curves was locally jagged and complex, which effectively expanded the dimensionality of the neural representation space and in turn facilitated a high-precision representation of AHV. Therefore, the encoding strategy for HD and AHV likely integrates characteristics of both dense and sparse coding schemes to achieve a balance between preserving specificity for individual features and maintaining their interdependency nature, marking a significant departure from the encoding of independent features and thus advocating future research delving into the encoding strategies of interdependent features.

show abstract

“…culminates into high level representations of visual objects which allow for object recognition (DiCarlo et al, 2012 ). Additionally, common neural population codes have been shown to have highly nonlinear relationships, suggesting nonlinear transformation is a common feature of sensory processing (De & Chaudhuri, 2023 ). The results of the modeling analysis provide additional evidence for this conclusion, revealing the best predictive power across all global properties is achieved after several transformations of the time-frequency input, which ultimately highlights the critical role of information abstraction that underlies the processing hierarchy in these deep models (Kell & McDermott, 2019 ; Yamins & DiCarlo, 2016 ).…”

Section: Discussionmentioning

confidence: 99%

Preliminary Evidence for Global Properties in Human Listeners During Natural Auditory Scene Perception

McMullin,

Kumar,

Higgins

et al. 2024

Open Mind

View full text Add to dashboard Cite

Theories of auditory and visual scene analysis suggest the perception of scenes relies on the identification and segregation of objects within it, resembling a detail-oriented processing style. However, a more global process may occur while analyzing scenes, which has been evidenced in the visual domain. It is our understanding that a similar line of research has not been explored in the auditory domain; therefore, we evaluated the contributions of high-level global and low-level acoustic information to auditory scene perception. An additional aim was to increase the field’s ecological validity by using and making available a new collection of high-quality auditory scenes. Participants rated scenes on 8 global properties (e.g., open vs. enclosed) and an acoustic analysis evaluated which low-level features predicted the ratings. We submitted the acoustic measures and average ratings of the global properties to separate exploratory factor analyses (EFAs). The EFA of the acoustic measures revealed a seven-factor structure explaining 57% of the variance in the data, while the EFA of the global property measures revealed a two-factor structure explaining 64% of the variance in the data. Regression analyses revealed each global property was predicted by at least one acoustic variable (R2 = 0.33–0.87). These findings were extended using deep neural network models where we examined correlations between human ratings of global properties and deep embeddings of two computational models: an object-based model and a scene-based model. The results support that participants’ ratings are more strongly explained by a global analysis of the scene setting, though the relationship between scene perception and auditory perception is multifaceted, with differing correlation patterns evident between the two models. Taken together, our results provide evidence for the ability to perceive auditory scenes from a global perspective. Some of the acoustic measures predicted ratings of global scene perception, suggesting representations of auditory objects may be transformed through many stages of processing in the ventral auditory stream, similar to what has been proposed in the ventral visual stream. These findings and the open availability of our scene collection will make future studies on perception, attention, and memory for natural auditory scenes possible.

show abstract

Common population codes produce extremely nonlinear neural manifolds

Cited by 5 publications

References 93 publications

Phantom oscillations in principal component analysis

Phantom oscillations in principal component analysis

Encoding of interdependent features of head direction and angular head velocity in navigation

Preliminary Evidence for Global Properties in Human Listeners During Natural Auditory Scene Perception

Contact Info

Product

Resources

About