Skip to main content

Inter-individual variability in learning: an investigation of its determinants

Periodic Reporting for period 1 - LearningDeterminants (Inter-individual variability in learning: an investigation of its determinants)

Reporting period: 2015-05-01 to 2017-04-30

The primate brain displays a tremendous capacity to learn. Yet, not all individuals are equal when it comes to learning abilities. Where do inter-individual differences in learning stem from and what is their impact on the acquisition of essential skills in our societies – such as reading and writing?
The present project proposes to:
- investigate the impact of demographic, life-style and cognitive measures on different forms of learning in adults (Aim 1a) and specifically assess the impact of technology use (Aim 1b).
- complement this knowledge with brain imaging (MRI), to reveal the brain correlates of individual learning abilities (Aim 2)
- evaluate the impact of individual differences in learning (specifically, audio-visual learning) on the acquisition of reading in young children (Aim 3).
Working towards these objectives, we have developed an array of new learning tasks and have acquired data on wide populations of adults through both in-lab and online testing.
1. Design and exploitation of an auditory structure learning paradigm
In line with Aim 1a, we have designed a novel auditory paradigm, assessing variability in structure learning abilities and its relation with cognitive skills. We have assessed the behaviour of a wide population of adults on a large battery of cognitive tests, both in-lab (N=50 participants) and online through the use of Mechanical Turk (N=300).
2. Investigation of the links between learning and technology use
As established in Aim 1b, we have focused on the impact of technology use (i.e. action videogame play) on statistical learning in both visual and auditory domains, testing a population of 100 participants in the lab.
3. Design and exploitation of an audio-visual associative learning task
In order to study another aspect of learning – namely audio-visual associative learning (Aim 1) – and its relations with reading, we have created a new learning paradigm that mimics phoneme-grapheme mapping. Difficulty has been finely adapted to our target population (young adults), allowing learning yet with inter-individual variability being observed.
We have acquired in-lab data on this task (N=40), measuring both learning and generalization to novel associations. A battery of validated questionnaires and cognitive tasks, widely used in the Brain and Learning Lab, has also been employed in order to identify the determinants of successful learning.
Furthermore, in line with project Aim 2, the same paradigm has been successfully adapted to magnetic resonance imaging.
Finally, consistently with project Aim 3, we have adapted the same task to use on a paediatric population, simplifying task demands as well as implementing child-friendly graphics and story-line.
The work described above has allowed us to establish that inter-individual variability in different types of learning (in particular in: structure learning, statistical learning and audio-visual associative learning) can be observed in adults, even in the context of short (maximum 1 hour) laboratory learning tasks. In the case of structure learning, we have replicated previous findings from the host lab, confirming that in our auditory paradigm – similarly to what was reported in a visuo-spatial version –, only 30% of the population displays the kind of structure learning we are after, in the allocated time.
Our experiments have also enabled us to start digging into the factors – be they demographic, related to life-style or cognitive –, that may be linked to variability in learning. One of the most relevant results is that, thanks to the completion of Aim 1a, we have confirmed our hypothesis that structure learning is linked to attentional skills in the general adult population, replicating this finding in both the in-lab and online populations tested.
The outcomes of Aim 1b – investigating the impact of action video-gaming (and thus heightened attentional control abilities) on statistical learning – preliminarily suggest subtle advantages following action video game practice, a finding in line with many of the host lab previous work.

Thus far, we have designed tools to evaluate the determinants of inter-individual learning and used those tools to highlight important links between learning abilities and cognitive skills. A distinctive feature of these tools is the possibility to administer them via online platforms to a large and representative sample of the population. While cognitive testing via online platform has become routine, getting robust measures of learning and how it progresses is new. At the end of the return phase, we expect firstly to gain knowledge on the neural correlates of efficient learning, by taking Aim 2 to the final analyses stage. Secondly, by implementing the testing phase of Aim 3, i.e. assessing associative audio-visual learning performances in children and relating them to reading abilities, we will further our knowledge of the factors favouring success in reading acquisition.
By asking what underlies successful learning and specifically reading acquisition (a gateway to many other forms of learning), the project directly addresses crucial societal challenges for the EU in the educational field, namely reinforcing the learning of basic skills such as literacy, as well as making progress towards the design of intervention strategies targeted at poor achievers.
Example of inter-individual variability in learning. Each line is meant to represent a participant.