1: \begin{abstract}\quad

2: Neural encoding studies explore the relationships between 

3: measurements of neural activity and measurements of a 

4: behavior that is viewed as a response to that activity. 

5: The coupling between neural and behavioral measurements is typically imperfect and difficult to measure.

6: To enhance our ability to understand neural encoding relationships, 

7: we propose that a behavioral measurement may be decomposable 

8: as a sum of two latent components, such that the direct neural

9: influence and prediction is primarily localized to the component which

10: encodes temporal dependence. For this purpose,

11: we propose to use a non-separable Kronecker sum covariance model to

12: characterize the behavioral data as the sum of terms with exclusively

13: trial-wise, and exclusively temporal dependencies.  

14: We then utilize a corrected form of Lasso regression in combination with the nodewise regression approach for 

15: estimating the conditional independence relationships between and among

16: variables for each component of the behavioral data, 

17: where normality is necessarily assumed.

18: We provide the rate of convergence for estimating the precision

19: matrices associated with the temporal as well as spatial components in the Kronecker sum model. 

20: We illustrate our methods and theory using simulated data, 

21: and data from a neural encoding study of hawkmoth flight; 

22: we demonstrate that the neural encoding signal for hawkmoth wing

23: strokes is primarily  localized to a latent component with temporal

24: dependence, which is partially obscured by a second component with trial-wise dependencies.

25: \end{abstract}

26: