Choice and learning in a 3-dimensional environment: an investigation of brain and behaviour

Project Information

CL-3DE

Grant agreement ID: 659680

DOI

10.3030/659680

Project closed

EC signature date 28 April 2015

Start date 1 May 2015

End date 30 April 2017

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 185 076,00

EU contribution

€ 185 076,00

Coordinated by

CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRS
France

Periodic Reporting for period 1 - CL-3DE (Choice and learning in a 3-dimensional environment: an investigation of brain and behaviour)

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

Summary of the context and overall objectives of the project

Impulsivity, the tendency to act prematurely without foresight, is associated with most forms of drug-taking and risky forms of behaviour. In humans, impulsivity research is almost always performed with participants who make choices or actions in response to images on a 2D screen. However, the scenarios that participants experience in such experiments are unlike those we encounter the objects in the real world. As a consequence, it is still unclear how impulsivity is influenced by real-world factors. Moreover, we do not know how these factors are processed in the brain. By replacing 2D cues with 3D objects using virtual reality technology, we were able to explore whether proximity to rewarding cues influences impulsive choice and impulsive action (Figure 1). We then adapted our concept for use with a functional MRI scanner to examine the neural correlates of proximity-driven impulsivity.

The British Dietetic Association’s (BTA) campaign for “junk free checkouts” in supermarkets came about because impulsive checkout purchases are thought to be contributing to the growing problem of obesity, especially among children. Our findings should be used to increase awareness among the public about the relationship between proximity and impulsivity in their consumer choices.

We had two core research objectives. First, using an immersive virtual reality technique, we aimed to examine how proximity to rewarding cues impacts impulsivity in both our choices and actions. Second, by adapting this technique for use in a MRI brain scanner, we aimed to understand how brain networks related to impulsivity are impacted by proximity.

For our first objective, we found that impulsive action, but not impulsive choice, was impacted by proximity. Although, the frequency of impulsive choices was the same when objects were near compared to when they were far, we found that participants were less able to withhold responses when rewarding objects were near compared to when they were far. For our second objective, we conducted an fMRI study aiming to explore the neural mechanisms involved in our behavioural finding. Data were collected for 28 human participants and analysis is still underway.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Study 1: It has been theorized that physical proximity to reward can lead to increases in impulsivity, but there has been no systematic study to examine this question. Using the Oculus Rift, we developed two immersive tasks to examine whether distance to reward impacts on two types of impulsivity (Figure 2). 1) Impulsive choice, an increased preference for immediate reward over a larger reward in the future, was examined using a 3D intertemporal choice task. Here, choices for smaller, immediate appetitive rewards were presented as tangible objects in space (Figure 3). Crucially, these were presented at either near (30cm) or far (360cm) distances. 2) Impulsive action, an inability to withhold a response, was examined using a 3D Go/Nogo task. Here, cues relating to Go and Nogo trials were presented using the same tangible objects in space. Crucially, both types of trials were presented at either near (30cm) or far (360cm) distances (Figure 4).
1) Results showed that impulsivity in intertemporal choice was not modulated by proximity. Participants (n = 27) did not make more choices for immediate rewards that were near compared to far. 2) Results from Experiment 2 showed that impulsivity of action was significantly modulated by proximity. Participants (n = 27) were less able to withhold their responses when objects representing Nogo trials were near compared to when they were far (Figure 5). This effect was robust even though participants slowed their reaction times for Go trials.
Findings suggest that self-control may be more likely to be compromised in an environment in which tempting cues are proximal. From an evolutionary perspective, we interpret this tendency for participants to behave more impulsively around near cues as an adaptive bias. However, viewed within a modern consumer society context, such a bias could be viewed as deleterious.
These findings have been presented at conferences. In addition, a research manuscript is in preparation to be submitted to Nature Human Behaviour.

Study 2: To follow up on the behavioural finding from Study 1, that impulsive action increases with proximity, we performed a second study which aimed to determine the neural mechanisms underlying such an effect. In this second study, we adapted the 3D Go/Nogo for use in the MRI brain scanner.
Performance of inhibitory control engages a different network to impulsive choice and includes prominent involvement of the lateral prefrontal cortex along with pre-SMA. A fronto-striatal circuit is also implicated in Go/No-go performance. We hypothesised that striatal activity would be modulated by task near/far Go/No-go manipulation.
We introduced two novel conditions such that Go and No-go trials were presented at near and far distances (Figure 6). This was made possible through stereoscopic presentation of cues through use of a ProPixx 3D Projector combined with a DepthQ polarisation modulator.
For this experiment, we used a jittered rapid-event related design to assess Go/No-go performance for near and far distances. Both Go and No-go trials were presented as tangible-seeming boxes and were associated with particular amounts of chocolate.
A total of twenty-eight participants have performed this task. We are currently using AFNI to analyse neuroimaging data. Over the next six months, findings will be written up for report in an international journal.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

Through collaboration with NeuroImmersion, we have developed the first example of an immersive computerised suite of neuroeconomics tasks which can provide a deeper understanding of how people make decisions and actions in the real world.
Once cheaper, more mobile brain measurement techniques become available, we envisage a coupling with techniques such as ours that will provide a much more detailed and nuanced picture of how behaviour, brain and environment interact.
In order to realise our aims, we have had to work closely with engineers from NeuroImmersion who specialise in 3D graphic and game design. We envisage more demand for close collaborations between scientists and software engineers, particularly game designers.
During the project I have made efforts to engage with citizens across ages to communicate the general knowledge from the field as well as our specific findings. I was invited to the Researchers’ Night in Brussels, to talk to schoolchildren, parents and teachers about my research and run demonstrations (Figure 7). This was a highly successful and publicised event that was attended by 3100 people on the day (https://ec.europa.eu/research/mariecurieactions/20160926-science-is-wonderful_en). I have also engaged with older students at universities in Lyon where I have given talks to communicate our methods and findings (Figure 8).
In addition, I have attempted to engage with a broader audience by writing articles for a general audience. One of these was recently published in the website of French Conversation (link: https://theconversation.com/delivery-drones-swooping-down-to-prey-on-our-self-control-77994). This has since been read by almost 3000 people.

Figure 7

Figure 4

Figure 1

Figure 5

Figure 2

Figure 8

Figure 6

Figure 3