Calibration of internal representations vs. external information for the control of action

While giving a presentation, we can speak freely from memory, but with some danger of losing our train of thought. Alternatively, we can rely on notes but risk a dysfluent and stilted presentation. This is one example of a type of decision that we need to make almost constantly; namely whether to rely on internal representations – that is, our knowledge and memory – or on information from the environment to guide our actions in the world.

Suboptimal internal/external decisions can have important consequences. For example, previous work has shown that older adults have a strong tendency to search the environment for clues, even when they know perfectly well what to do next, which in turn leads to highly inefficient performance. Other vulnerable populations, such as individuals with attention deficit and hyperactivity disorder (ADHD), are likely to rely either too much or too little on external information and therefore can appear either distracted or hyper-focused.

Adequate balancing between information sources is particularly important when people interact with modern and emerging technologies. For example, the increasing salience of information-rich displays in newer-model cars can pose a serious distraction danger for individuals with a strong tendency to rely primarily on external information. Yet, little is known about how internal/external decisions are being made, what kind of information is used for such decisions, and what exactly explains the suboptimal pattern in certain subgroups, such as in older adults or individuals with ADHD.

The current project will use computational modeling and eye-tracking experiments to address this gap in the research literature. The long-term goal of this work is to inform technological solutions that support optimal internal/external information calibration, in particular in vulnerable populations.

The investigators use eye-tracking technology to monitor human participants in experimental situations that require performing one of several, simple tasks at any given point in time (e.g., responding to an arrow cue by indicating either its spatial location or the direction in which it is pointing). The currently most appropriate task can be determined either by utilizing the participant’s own knowledge (internal representations) or by examining simple reminders that indicate what to do next (external information).

In Aim 1, the focus is on aspects that can “irrationally” influence the internal/external decision process and that “pull” the decision process towards relying on the environment, even when there is no objective need to use external information. For example, preliminary results have shown that both the saliency of external prompts and how frequently an individual had inspected external signals in the recent past, strongly influence the current use of information in the environment.

Aim 2 focuses on situations in which there is an optimal balance between relying on external or internal information that participants need to try to achieve. The goal here is to determine to what degree external/internal decisions can be explained by a simple, rational model, and if not, how such a model would have to be amended in order to explain participants’ behavior.

For example, preliminary evidence suggests that the subjective cost involved in making internal/external decisions itself is a factor that can steer individuals towards suboptimal patterns of information use. Aim 3 combines the experimental paradigms from the previous aims to characterize internal/external decisions within a larger, heterogenous sample of young and older adults, as well as individuals diagnosed with ADHD.

The project also includes an education and outreach component directed towards increasing accessibility of STEM education for students with ADHD. Students from the University of Oregon’s School of Journalism will develop animated material about the nature of ADHD that will be tested in the University’s Accessible Education Center. Combined, this work will generate new theoretical and empirical knowledge about how individuals use internal versus external sources of information and will lead to a better understanding of suboptimal information use in vulnerable populations.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.