Georgopoulos and Carpenter publication in Current Opinion in Neurobiology

An article entitled "Coding of movements in the motor cortex" by Brain Sciences Center members Adam Carpenter and Apostolos Georgopoulos appears in the August 2015 Current Opinion in Neurobiology.



  • Defines coding of movements in motor cortex and traces its origin and history.
  • Focuses on the radical shift in 1980, from muscles/joints to extrapersonal space.
  • Traces the path of discovery of evolving neural movement trajectory in the 1980s.
  • Highlights the relations between coding of movement and neuroprosthetic applications.
  • Focuses on local inhibitory mechanisms as key factor for movement accuracy and speed.

Coding of movements in the motor cortex

The issue of coding of movement in the motor cortex has recently acquired special significance due to its fundamental importance in neuroprosthetic applications. The challenge of controlling a prosthetic arm by processed motor cortical activity has opened a new era of research in applied medicine but has also provided an 'acid test' for hypotheses regarding coding of movement in the motor cortex. The successful decoding of movement information from the activity of motor cortical cells using their directional tuning and population coding has propelled successful neuroprosthetic applications and, at the same time, asserted the utility of those early discoveries, dating back to the early 1980s.


Below:Schematic diagrams to illustrate the hypothesis of directional accuracy via a variably tuned inhibitory drive. Figure 3 (left): Strong inhibitory drive leads to accurate and slower movement by reducing the directional tuning width and producing an accurate and short population vector. Red and black terminals indicate excitatory and inhibitory synapses, respectively. P, pyramidal cell; I, inhibitory interneuron. Figure 4 (right): Weak inhibitory drive leads to less accurate and faster movement by increasing the directional tuning width and producing a less accurate and longer population vector.




"An information theory analysis of spatial decisions in cognitive development", Nicole Scott, Maria Sera and Apostolos Georgopoulos, Frontiers In Neuroscience - Decision Neuroscience, Feb. 4 2015

Performance in a cognitive task can be considered as the outcome of a decision-making process operating across various knowledge domains or aspects of a single domain. Therefore, an analysis of these decisions in various tasks can shed light on the interplay and integration of these domains (or elements within a single domain) as they are associated with specific task characteristics. In this study, we applied an information theoretic approach to assess quantitatively the gain of knowledge across various elements of the cognitive domain of spatial, relational knowledge, as a function of development.

Abstract continued...
Specifically, we examined changing spatial relational knowledge from ages 5 to 10 years. Our analyses consisted of a two-step process. First, we performed a hierarchical clustering analysis on the decisions made in 16 different tasks of spatial relational knowledge to determine which tasks were performed similarly at each age group as well as to discover how the tasks clustered together. We next used two measures of entropy to capture the gradual emergence of order in the development of relational knowledge. These measures of "cognitive entropy" were defined based on two independent aspects of chunking, namely (1) the number of clusters formed at each age group, and (2) the distribution of tasks across the clusters. We found that both measures of entropy decreased with age in a quadratic fashion and were positively and linearly correlated. The decrease in entropy and, therefore, gain of information during development was accompanied by improved performance. These results document, for the first time, the orderly and progressively structured "chunking" of decisions across the development of spatial relational reasoning and quantify this gain within a formal information-theoretic framework.


Peka ChristovaPeka Christova's paper "Innovations in Resting-state fMRI and MEG" is 'in press' at the Asian Journal of Physics, Vol. 23, No 5 (2014) 849-857.

Functional magnetic resonance imaging (fMRI) and magneto-encephalography (MEG) are modern imaging techniques that rely on physical phenomena to record brain activity. Both methods are non-invasive and require sophisticated equipment and recording conditions. The biological phenomena upon which they are based are different: fMRI measures the Blood Oxygenation Level-Dependent (BOLD) level, which reflects local hemodynamic changes, whereas MEG directly measures integrated local synaptic activity. The superior temporal resolution of MEG allows the assessment of short temporal interaction between the brain regions. Such assessments can apply to healthy subjects and patients. On the other hand, the superior spatial resolution of fMRI allows the precise localization of brain structures with a detailed inside view of brain connectivity patterns. The resting-state recordings of BOLD and MEG signals were analyzed as time series of each voxel/sensor. To estimate the true correlations between them, the prewhitened time series, called innovations, were used.



Updated February 25, 2015