Roland Lab – University of Copenhagen

Forward this page to a friend Resize Print Bookmark and Share

CNS > Research > Section for Integrative Neuroscience > Roland Lab

Roland Lab


Neuroscience does not have, as physics does, a standard model that serves as a conceptual structure in which gaps of knowledge and inconsistencies can be isolated and serve as impetus for experiments, technological improvements or elaborate calculations. In other words, we do not understand how mammalian brains work. Mammalian brains produce perception, thoughts and behavior over a wide spectrum ranging from a reflex to theory of relativity. So, far, explanations on how brains may produce percepts, thoughts or behavior relied on experimental results showing temporal dynamics of spiking activities and brain models explaining aspects of temporal dynamics of spiking and membrane conductances. For perspectives on which aspects a common brain theory should include see the special issue of Neuron on “How does the brain work?”, initiated by this unit.
Neuron, June 7, 2017

Neurons causally affect their target neurons by sending action potentials (spikes) through their axons thereby altering the membrane currents (MC) of their target neurons. Conversely, exposed to sufficiently strong excitatory membrane currents, a neuron emits a strong and fast membrane current, an action potential or spike. These two causal relations form the basic machinery of the brain (for an example see Movie 1 below). Previous explanations of brain mechanics assumed particular neurons and particular subpopulations of neurons representing particular features or objects in the physical surround, or assumed specific relations between items or features in the surround and spiking patterns, spiking codes, spiking statistics, current or spiking synchronicity, noise, and internal models etc. in the brain. Our main hypothesis is that space-time dynamics of action potentials and membrane currents drive a mammalian brain to produce perception in less than 150 ms, and thoughts and behavior in less than 1500 ms.


Our group examines the space-time dynamics in the cerebral cortex, i.e. the progression of membrane current changes, local field potential changes and spiking in the space made up of the cortical network of neurons, or in state space, with the purpose of finding principles of cortical mechanics (for an example see Movie 2 below).

We also examine the dynamics of action potential sequences from single excitatory and single inhibitory neurons in the cerebral cortex of epilepsy patients, in the awake state, during simple task performance, during sleep, prior to epileptic auras and seizures, during seizures and in the post-ictal phase, with the purposes of understanding the mechanisms of perception, planning of movement and distinguish abnormal dynamical evolutions from normal.


Roland, P.E. (2017) Space-time dynamics of membrane currents evolve to shape excitation, spikjing , and inhibition in the cortex at small and large scales. Neuron 94: 934-942.

Roland, P.E., Bonde, L. H.,
Forsberg, L., Harvey, M (2017) Breaking the excitation-inhibition balance makes the cortical network’s space-time dynamics distinguish simple visual scenes. Front. Syst. Neurosci. 11:14 doi: 10.3389/fnsys. 2017.00014.

Forsberg LE, Bonde LH, Harvey MA, Roland PE (2016) The second spiking threshold: Dynamics of laminar network spiking in the visual cortex. Front. Syst. Neurosci. Doi: 10.3389/fnsys.2016.00065.

Huys, R., Jirsa, V.K., Darokhan, Z., Valentiniene, S., Roland, P.E. (2015). Visually evoked spiking evolves while spontaneous ongoing dynamics persist. Front. Syst. Neurosci. Doi: 10.3389/fnsys 2015 00183.

Harvey, M.A., Valentiniene, S. ,and Roland, P.E. (2009). Cortical membrane potential dynamics and laminar firing during object motion. Front. Syst. Neurosci. 3:7. doi:10.3389/neuro.06.007.2009

Overall publication metrics: Number of papers: 140, H-index: 62; i10 index: 111.

Movie 1 Predictive depolarization and spiking. One animal exposed to a bar moving down from the peripheral field of view starting at 0 ms. The retina is stationary. Note that the bar then is mapped as moving excitation over the cortex. However, at 104 ms the neurons in area s 19/21 compute an excitation far ahead of the bar mapping. After feedback to areas 17/18 this repeats here. The black holes show the electrode penetration sites along the border between areas 17 and 18 corresponding to the vertical meridian. When the spiking at any layer of the cortex becomes statistically significant (p < 0.01) the hole turns white. Note the mapping of the future bar trajectory when the bar representation on the cortex has reached the left white arrow (155 ms). Note also how the object mapping, defined by the hot spot in area 17/18 actually follows the cortical route predicted already at 160 ms. Animal 410. (From Harvey et al. 2009).

Movie 2 State-space dynamics of spiking of 50 single trial trajectories in response to a bar moving down from center of field of view. Several projections of state space shown. The projections below the diagonal are redundant. The figures along the diagonal show the projections of the single trials for the first 4 principal components. The total variance accounted for by the two principal components is shown on the top of each state space projection in 2 dimensions. The trajectory of a single trial is represented by a red dot shoving the instantaneous position in state space. The dot has a tail showing where the trHent videoen i høj ajectory was up to 20 ms prior to actual time. The yellow dot is the instantaneous center of gravity. Electrode penetration in the cortex mapping the CFOV (center of field of view) (From Forsberg et al. 2016).