How reliable are International assessment programs?

Outcomes from international student, such as the OECD’s Program for International Student Assessment (PISA), are frequently used to guide educational policies. For example, the recent OECD Education at a Glance report highlighted the importance of quality school principals. Because of their wide audience and the regard in which they are held, it’s important that these findings are confirmed.

In this paper, Martin Carnoy and colleagues attempt to gauge the relative importance of teacher quality, a factor generally considered highly influential in student outcomes. They analysed the performance of the same Russian students in the mathematics portion of the PISA 2012 assessment and the 2011 TIMSS (Trends in International Mathematics and Science Survey), as well as details of their relevant teachers and curriculum. This is a much more detailed analysis of mathematics performance in a single population, as opposed to the cross-sectional data taken by a single PISA study, for example. The authors find that for the Russian group studied, the impact of teachers was more modest than suggested by the PISA report. Nevertheless, they stress that teacher quality is an important component of quality education, only less so than suggested by a cross-sectional study design as used in PISA.

Carnoy et al. (2016) Revisiting the relationship between international assessment outcomes and educational production: Evidence from a longitudinal PISA-TIMSS sample. American Educational Research Journal 53(4): 1054-1085

Confusing memories: how we distinguish between similar events

Our hippocampus is where our autobiographical memories live – like what we had for breakfast this morning, or which parking bay our car is in. Sometimes, these memories can be very similar to each other; maybe the bay we parked in yesterday was one level down from the one we used today. How do we distinguish these memories?

Models and experiments in animals suggest that a particular part of the hippocampus, called the dentate gyrus, is responsible for separating similar memories. To some extent, this has been confirmed in humans. Neuroimaging experiments have shown that the dentate gyrus is active when discriminating two similar memories, and patients with hippocampal lesions have deficits in discrimination. But these experiments don’t tell the full story: the neuroimaging data don’t allow causation to be inferred, and the lesion experiments have so far only included patients with damage to large portions of the hippocampus, not specific for the dentate gyrus.

In this investigation, a 54-year-old man with damage limited almost entirely to his dentate gyrus was studied. As expected, the man had difficulty correctly determining which items were new if they were similar to items he had seen before – he thought he had seen them before even though he hadn’t. This deficit fits with a role for his damaged dentate gyrus in distinguishing similar items in memory.

Baker et al. (2016) The human dentate gyrus plays a necessary role in discriminating new memories. Current Biology (Advance Online Publication)

Special circuits for learning by watching

One of the major ways we learn is by watching others. This is true both in the real world and in the classroom, where learning by observation (and instruction) normally precedes learning by doing. What are the differences in how our brains handle observational and self-experienced learning?

To test this, Itzhak Fried and colleagues recorded individual neurons in the brains of epilepsy patients who were undergoing surgeries to identify the source of their seizures. They looked at three brain areas, comparing the neural signatures of error and reward learning under two situations: when the subject played a gambling task themselves, or when they watched somebody else play. Two of the brain regions, the amygdala and rostromedial prefrontal cortex, showed identical activity patterns regardless of who did the gambling. In contrast, the third region (the rostral anterior cingulate cortex, or rACC) was far more active during observational learning.

The results identify one of the neural differences between observational and experience-based learning – the heavy involvement of the rACC for observational learning. The rACC is thought to be important for emotional components of error detection, but why it is more strongly activated during observation than performance is not clear.

Hill et al. (2016) Observational learning computations in neurons of the human anterior cingulate cortex. Nature Communications, 7: 12722

Individual differences in reading ability coincide with altered brain connectivity

Differences in reading ability are apparent in childhood and persist into adulthood, with reading difficulties such as dyslexia existing at the low end of a reading ability continuum. A major cause of reading problems is difficulty in processing sounds. However, other factors, such as attention and visual processing, also play a role. In this study, researchers investigated how a brain area involved in sound processing, the temporoparietal junction (TPJ), interacted with regions needed for other tasks, such as attention or visual processing. This connectivity data, measured while the subjects rested with their eyes closed, was then compared on a case-by-case basis with their reading performance.

The authors were able to show that reading ability correlated with how well the TPJ network connected with other networks. This included seeing an increase in connectivity with networks important for attention, which likely reflects compensation for poor reading by increased focus. What is surprising is that the increased connectivity between attention and sound processing regions was measured when the individual was resting with their eyes closed, not reading. This suggests that in poor readers, even when they aren’t reading, increased attentional resources may be diverted to reading regions of the brain.

Achal et al. (2016) Individual differences in adult reading are associated with left temporo-parietal to dorsal striatal functional connectivity. Cerebral Cortex 26(10):4069-4081