Hebbian activity-dependent plasticity in white matter

Abstract

Synaptic plasticity is required for learning and follows Hebb's rule, the computational principle underpinning associative learning. In recent years, a complementary type of brain plasticity has been identified in myelinated axons, which make up the majority of brain's white matter. Like synaptic plasticity, myelin plasticity is required for learning, but it is unclear whether it is Hebbian or whether it follows different rules. Here, we provide evidence that white matter plasticity operates following Hebb's rule in humans. Across two experiments, we find that co-stimulating cortical areas to induce Hebbian plasticity leads to relative increases in cortical excitability and associated increases in a myelin marker within the stimulated fiber bundle. We conclude that Hebbian plasticity extends beyond synaptic changes and can be observed in human white matter fibers.

Keywords: CP: Neuroscience; Hebbian plasticity; action reprogramming; brain plasticity; brain stimulation; magnetic resonance imaging; myelin; myelin plasticity; white matter.

Graphical abstract

Figure 1

Inducing Hebbian plasticity in the human brain (A) Summary of experimental design, using two cohorts to establish effects of Hebbian stimulation on brain microstructure. Study 1 (n = 19) included the Hebbian condition only. In study 2, a different set of individuals were randomized to receive either Hebbian (n = 18) or non-Hebbian (n = 18) stimulation. (B) Diagram of the Hebbian (active) and non-Hebbian (control) conditions used in the experiments. Both stimulation protocols are matched for duration, intensity, and coil location but differ in the relative timing of the stimulation pulses, with the Hebbian condition aiming to mimic the timing of synaptic plasticity inductions used in vitro. (C) Longitudinal effects of Hebbian-plasticity induction on cortical physiology. Each dot in the graph represents the normalized change in cortical excitability (as measured by the SI_1mV metric) for one subject. The SI_1mV measure was collected in an exploratory manner in the last 7 participants of study 1 and in all participants of study 2 to confirm the presence of longitudinal effects.

Figure 2

Microstructural plasticity induced by Hebbian stimulation (A) Results from a whole-brain analysis identify a cluster where changes in MT values correlate with changes in cortical excitability in the Hebbian condition significantly more than they do in the non-Hebbian condition. (B) The significant MT cluster identified by the whole-brain analysis (red) overlaps with stimulation sites in the gray matter and with the stimulated fiber tract in the white matter (blue). (C and E) Scatterplots of data underlying the significant cluster. For the Hebbian condition, participants with greater increases in excitability (more negative physiological change score) show greater increases in MT. Each data point is a single participant; scatterplots (with line of best fit and 95% confidence bands) are presented for post-hoc visualization of the correlations rather than for statistical inference. (D and F) Tracking of stimulation sites via neuronavigation allows us to estimate the location of cortical stimulation sites and to reconstruct the stimulated fiber bundle in white matter.

Figure 3

Hebbian stimulation induces anatomically relevant changes in action reprogramming (A) Schematic of the action-reprogramming task used, based on (Neubert et al., 2010), probing both action execution (stay trials) and action reprogramming (switch trials). (B) Premotor-to-motor circuits are involved in action reprogramming, as exemplified by a meta-analysis of action-reprogramming task fMRI studies. (C) Reaction times during the task increase in switch trials (when the cue changes) compared with stay trials (while the cue remains the same) in all studies. (D) Summary of experimental design, testing the effects of Hebbian stimulation on action reprogramming in two cohorts. (E) Longitudinal effects of Hebbian-plasticity induction on action-reprogramming behavior. Each dot in the graph represents the normalized change in switch-trial reaction time for one subject.

Figure 4

Large-scale compensatory changes in resting-state connectivity induced by Hebbian stimulation (A) Results from a whole-brain analysis identify clusters where connectivity changes correlate with changes in cortical excitability in the Hebbian condition significantly more than they do in the non-Hebbian condition. (B and C) Scatterplots of data underlying the significant cluster. Each data point is a single participant; scatterplots (with line of best fit and 95% confidence bands) are presented for post-hoc visualization of the correlations rather than for statistical inference.