Optimizing perception across the adult lifespan

Abstract

While our sensors allow us to take up information on the world around us, perception is the process by which we make sense of it. Perception is one of the most essential functions of the human mind as it connects us with the world and allows us to interact with it – ensuring well-being and independence. It is intimately tied to other processes, such as motor control and memory. While perception seems trivial, the underlying processes are quite complex. The information obtained from our senses is inherently ambiguous and uncertain. Perception is understood as a probabilistic inference process that seeks to reduce uncertainty in order to provide us with a valid interpretation of our surroundings. During this process, predictions play a central role. Prior and sensory information are considered to be fused (based on their relative reliability) to infer the most probable state of the world. Aging introduces a particular challenge to perception as it gives rise to even more uncertainty: As we grow older, internal noise in nearly all sensory systems increases while our tolerance for external noise decreases. In parallel, many cognitive abilities – such as working memory or processing speed – are subject to pronounced decline. However, one particular strength of growing older is that we gain more and more knowledge about ourselves and the world around us. This knowledge may provide a powerful resource to adapt to current challenges and could be used to improve our predictions. But in order to be beneficial, sensory and predictive signals need to be adequately balanced – in theory, predictions should become more important with age. Across two different studies, I could show that aging increases the reliance on predictions, which has consequences for both motor control and memory processes. The first study examined how age and cognitive processes modulate tactile suppression during reaching. Tactile suppression was about three times stronger in older adults compared to younger adults, indicating greater reliance on sensorimotor predictions with age. Furthermore, increased task demands due to the introduction of an additional memory task overall led to increased tactile suppression but did not modulate the age effect. Across age groups, stronger tactile suppression effects were associated with lower cognitive control capacities. The second study focused on the impact of prior knowledge on object memory. I investigated whether older and younger adults’ memory performance for objects embedded in real-world scenes were differentially affected by object-scene inconsistencies (e.g. ketchup in the shower). Objectscene inconsistencies were beneficial for memory performance in both age groups. However, this memory advantage was attenuated in older adults. Additionally, older adults showed an enhanced congruency-bias when asked to match recognized objects to their original context. The findings from this study highlight the role of predictive processes for memory performance and reveal slight disadvantages when predictions are violated. While greater reliance on predictions may overall be an adaptive mechanism that helps us to improve our performance across different tasks, the decisions we make may still be wrong at times. Luckily, our decisions are usually accompanied by a subjective feeling of confidence. Confidence is an online process that monitors and controls our decisions and informs us whether we should trust or doubt them. It is highly important for successful behavior. In two studies, I investigated how perceptual confidence is affected by age and increased task demands. In the first study, I could show that both younger and older adults can adequately distinguish between correct and incorrect perceptual decisions. Higher confidence was associated with better performance. However, on average, this ability was reduced in older adults compared to younger adults. Furthermore, individual differences in confidence efficiency and cognitive control capacities were closely related. The second study challenged metaperceptual abilities in a sample of younger adults and revealed that confidence comparisons across visual and tactile decisions are possible without any major costs in confidence efficiency or response times compared to confidence judgments made within the same modality. The study provided evidence that confidence is stored in an abstract, modality-independent format. It remains to be clarified how this ability is affected by age. In contrast to the previous study, though, confidence efficiency and cognitive control capacities were not correlated – most likely due to variance restrictions in an age-homogeneous sample. Taken together, the four studies presented in this thesis contribute to our understanding on how different information is valued in response to individual challenges and current task demands. They further provide evidence that valuation processes are largely preserved with age and may serve as adaptive mechanisms to optimize performance in the face of pronounced sensory and cognitive changes. However, the efficiency of these mechanisms seems to rely on the availability of cognitive control resources.