Last month we had the opportunity to share ideas regarding neuropsychological rehabilitation and new technologies with a lot of professionals at the INS Mid-Year Meeting in Amsterdam. It was a very interesting meeting in which we could attend to worldwide experts´ conferences in our field, and meet other professionals. They had a broad range of profiles: neuropsychology, neuroscience, and neurology, as well as occupational therapy or social work.
All of them considered interesting some advances we took into account when we created the platform: customization, data tracking, the possibility to use it as an assessment tool (or even an experimental tool), levels, possibility to change parameters in the activities…
Even so, they all had similar concerns about our tool. So we think it´s fair to detail some aspects we deal with when we decided to create NeuronUP. And when we talk about “some aspects”, we mean “ecological validity”, which was the main issue we were asked for.
Some may say it´s too sad to get old or injured and be confronted with a reality you may never been confronted to: computers. It´s frustrating and it may impact in motivational aspects, a powerful issue for functional recovery.
WE THINK THE SAME
We don´t believe in computer-based rehabilitation when it might produce more harm than good.
Objectives, as well as formats of our rehabilitation schedule/program must be adjusted to our patient´s cognitive profile and to its preferences and motivations, when possible. The goal is not only to reach a certain level in a set of constructs, but to be functional. These are two main concepts in neuropsychological rehabilitation (which covers a wide range of interventions). According to Burgess et al. (2006) we could include a third component, operations, to obtain the myriad of levels of analysis in neuropsychology.
So, we have constructs (theoretical, inferred, such as “working memory”), operations (realities that correlate manipulated variables of a task and changes in BOLD signal, such as mental rotations), and functions (directly observable, the result of the interaction between a subject´s brain and a goal, i.e. “social behavior”).
These three concepts had a lot to do with the different tasks we use to rehabilitate. Kvavilashvili & Ellis (2004) detail three levels in cognitive rehabilitation (see the graphic below). They also point to the necessity of making representative and predictive tasks. Representative means that the task we are using in rehabilitation has to do with (or it´s overlapped to) a real-task. Predictive means that the results of the performance in the “lab”-tasks are generalized to real tasks. We already know the solution in the field of neuropsychological assessment to these issues: design nearly-real tasks, not simplified versions (or what we think it´s a simplified version) of a real task.
“Why don´t we design the same tasks for neuropsychological rehabilitation?” Some may say that could impact efficacy, that the use of almost-real tasks impacts validity. The truth is that we don´t believe in efficacy in neuropsychological rehabilitation. Efficacy is a concept that comes from situations where you can control almost all variables: Physics, Chemistry, Pharmacy or Medicine. We need practical results accepting that neuropsychological rehabilitation still have to be addressed in a deeper analysis. Moreover, neuropsychological rehabilitation is an active process that requires the participation of the patient to get the goals (so it is impossible to design a blind group where people ignore If they are “receiving” the treatment or not). We agree with the need to check out the clinical impact of the tasks we design. But when we consider this, effectiveness is in our mind. We shouldn´t forget that a rehabilitation tool doesn´t have the same goal that an assessment tool. And a rehabilitation tool is not a test. A test may be used in one, two, three ways… as it has a standardized protocol (or more) and population scales. A rehabilitation tool should be enough flexible to be used in as many ways as possible, with the possibility to be adjusted to our patient´s cognitive profile and strategies.
If we take a look to what represents ecological validity in neuropsychological assessment, we can see that the Zoo Test (Alderman, 1996) or the Multiple Errands Test (Shallice & Burgess, 1991) have acceptable inter-rater reliabilities and a powerful psychometrical value. “Real-world model” tasks may also have particularly good generalisability. While the results of other experimental tasks with a widespread use still have to be replicated, especially in what concerns to generalisability.
It´s necessary to point out that neuropsychological rehabilitation is not only composed of ecological tasks. We know that basic tasks (“lab-tasks”) and basic strategies can be used depending on the moment, the type of injury, the age of the patient, its cognitive reserve or the rehabilitation potential, among others. And they can be effective, depending on the strategy we have set out.
So we have different analysis levels in neuropsychological rehabilitation, with different goals and activities associated. Our competence, as professionals, is to decide which type of activity is better for the patient and which one adapts better to his/her motivations. And videogames can be a useful option.