Week 9 model answer
Snowball et al (2013) built on work that transcranial random noise stimulation (TRNS) could enhance neuroplasticity, and on work which found that dorsolaternal prefrontal cortex (DLPFC) is a key brain area for arithmetic. A key question for any study of learning or neuroplasticity is whether an effects of an intervention are long term.
25 paid participants were divided into two groups. One group received TRNS over the DLPFC, the other group was treated identically except no current was passed through the equipment (a sham control). The participants and experimenters were blind to which conditions participants were in. Over five days participants practiced interleaved blocks of calculation and drilled arithmetic tasks, from which performance measures were recorded (error and reaction time) and learning curves derived. Near infrared spectroscopy (NIRS) was used to record PFC hemodynamic response on the first and last day of training. Six months later some participants were brought back and re-tested, again with NIRS.
The TRNS condition learnt faster than the sham condition, for both calculation and drill problems. This generalised to new problems for the calculation problems only. The TRNS group showed differences in hemodynamic response, specifically for the peak time of all three assessed components (but independent of day of training).
After six months delay, there was no significant benefit of training on either the drill or calculation problems (p=0.08 for the calculation condition, if a two-tailed test used). A significant correlation existed between the NIRS peak time measures and calculation RTs at this point, across both groups (Figure 4B).
CONCLUSIONS, IMPLICATIONS AND CRITICISMS
This is a technically impressive piece of research which suggests that learning can be directly enhanced by brain stimulation. Although a number of important controls are done, questions still remain about the mechanism, generalisability and longevity of any effect.
The use of NIRS acts a manipulation check, showing that TRNS is having an effect on brain function (compare with doubts about TDCS raised by Horvath, 2014).
The big weakness of the study is the small numbers are non-significant effect at six months re-test (which they report as significant, but only for a one tailed test. It is perfectly reasonable to expect the long term effect of TRNS to be either positive or negative, so a one-tailed test is not justified).
Additional controls which could have been performed are
1. Using a combined measure of error rate and reaction time to control for speed-accuracy trade-offs. Benefits of TRNS are shown on reaction times only, and it is reported that there were no significance differences on error rates - but an absence of evidence is not the same as evidence of absence. Error rates could still have varied systematically, because participants were making a speed-accuracy trade off, but the only significant differences show up on the RT measures. For this reason, it would be better to analyse a variable which combined RT and accuracy.
2. Systemmatically testing of participant knowledge of condition. The authors state that participants did not know which condition they were in (TRNS or sham) but no test of this was performed. Although they were asked if they knew this could be contaminated by reporting bias (participants might be reluctant to guess). A better test would be to force participants to guess and to test if they were able to do this at above chance levels.
3. Specificity of effect.
The mechanism of action – what exact brain function – is not clear. Although a control experiment stimulating the parietal cortex was done, it is still not definitive that the effect was because of DLPFC stimulation, nor is it clear what aspect of DLPFC function was affected. Related to this, before any wider deployment of the technique in an education testing further testing would be needed to check that TRNS didn’t impair other functions not measured here. This is a reasonable concern because competitive interactions are a general principle of brain development, so it might be that an overactive DLPFC reduced development.
Horvath, J. C., Forte, J. D., & Carter, O. (2014). Evidence that transcranial direct current stimulation (tDCS) Generates little-to-no reliable neurophysiologic effect beyond MEP amplitude modulation in healthy Human subjects: A systematic review. Neuropsychologia.
Snowball, A., Tachtsidis, I., Popescu, T., Thompson, J., Delazer, M., Zamarian, L., Zhu, T., Cohen Kadosh, R. (2013). Long-Term Enhancement of Brain Function and Cognition Using Cognitive Training and Brain Stimulation. Current Biology, 23, 987–992. doi:10.1016/j.cub.2013.04.045Ed