After a person suffers a traumatic brain injury, they will usually be assessed via a number of paper-and-pencil tests and desk-based performance measures to see if they have acquired any cognitive impairments. A problem with this approach is that such tasks arguably lack ecological validity – that is, they don't realistically resemble the kind of tasks people face in their everyday lives. For instance, neuropsychological tasks typically tap one cognitive function at a time, whereas real-life challenges engage multiple cognitive faculties all at once. The ideal would be to take brain-injury patients out into the real world to test their performance, but there are problems with that too. The patient may be placed at risk (for example when they first try crossing roads or driving again) and by its very nature, the real world is obviously impossible to control – there would be an infinite number of variables affecting their performance.

The solution could be virtual reality (VR). Fully immersive, three-dimensional VR can be used to safely mimic some of the complexity of the real world, whilst also allowing the kind of control and repeatability needed for psychological testing. Now Robert Matheis (Fairleigh Dickinson University) and colleagues have carried out an exploratory study to investigate the practicalities of this approach for testing memory performance.

Using a VR headset connected to a laptop, Matheis and colleagues tested the memory performance of 20 traumatic head injury patients and 20 healthy controls within a virtual office environment. Most of the patients had suffered their injuries in traffic accidents. The participants had to learn 16 target items located around the virtual office. They were allowed up to 12 visits to the virtual environment to learn the 16 items. Thirty minutes after the learning phase, their memory for the items and their locations was tested; they were also tested again 24 hours later. The participants also performed a raft of conventional neuropsychological tests including the California Verbal Learning Test. 

The VR memory task certainly succeeded in distinguishing some of the patients from controls – 25 per cent of the patients were unable to learn all 16 items in the time allowed, whereas all the controls managed this. However, the recall and recognition tests at 30 minutes and 24 hours failed to find any difference in ability between the remaining 16 patients (those who were initially able to learn all 16 items) and the control participants. That's despite the fact the controls significantly outperformed these patients on the California Verbal Learning Test. This suggests the VR memory task was easier, which is probably no surprise given that, just as in real life, the VR task allows participants to use visual and environmental cues to help them remember the items. The researchers said: "Thus while testing within a virtual environment can afford the examinee advantages, these advantages are quite representative of the domain of functioning to which we desire our assessment to generalise".

In support of the VR test's construct validity (that it's measuring what it claims to be),  there was a large deal of correlation between each participant's performance on the 30 minute recall test and their performance on the California Verbal Learning Test (49 per cent of variance was shared). Commenting on this, the researchers said: "it appears the VR office may be tapping memory abilities similar to the established CVLT, but that the two instruments clearly also yield unique data that together may form a more accurate picture of an examinee's real-world performance".

Web Directions:

Matheis, R.J.,  Schultheis, M.T., Tiersky, L.A., DeLuca, J., Millis, S.R. & Rizzo, A. (2007). Is learning and memory different in a virtual environment? The Clinical Neuropsychologist, 21, 146–161. http://dx.doi.org/10.1080/13854040601100668

The Integrated Media Systems Centre at the University of Southern California, one of the participating laboratories in this study. http://imsc.usc.edu/

Information on the California Verbal Learning Test: http://www.memorylossonline.com/glossary/californiaverballearningtest.html