Language models, like humans, show content effects on reasoning tasks.

reasoning is a key ability for an intelligent system. Large language models (LMs) achieve above-chance performance on abstract reasoning tasks but exhibit many imperfections. However, human abstract reasoning is also imperfect. Human reasoning is affected by our real-world knowledge and beliefs, and shows notable "content effects"; humans reason more reliably when the semantic content of a problem supports the correct logical inferences. These content-entangled reasoning patterns are central to debates about the fundamental nature of human intelligence. Here, we investigate whether language models-whose prior expectations capture some aspects of human knowledge-similarly mix content into their answers to logic problems. We explored this question across three logical reasoning tasks: natural language inference, judging the logical validity of syllogisms, and the Wason selection task. We evaluate state of the art LMs, as well as humans, and find that the LMs reflect many of the same qualitative human patterns on these tasks-like humans, models answer more accurately when the semantic content of a task supports the logical inferences. These parallels are reflected in accuracy patterns, and in some lower-level features like the relationship between LM confidence over possible answers and human response times. However, in some cases the humans and models behave differently-particularly on the Wason task, where humans perform much worse than large models, and exhibit a distinct error pattern. Our findings have implications for understanding possible contributors to these human cognitive effects, as well as the factors that influence language model performance.

An implicit memory of errors limits human sensorimotor adaptation.

During extended motor adaptation, learning appears to saturate despite persistence of residual errors. This adaptation limit is not fixed but varies with perturbation variance; when variance is high, residual errors become larger. These changes in total adaptation could relate to either implicit or explicit learning systems. Here, we found that when adaptation relied solely on the explicit system, residual errors disappeared and learning was unaltered by perturbation variability. In contrast, when learning depended entirely, or in part, on implicit learning, residual errors reappeared. Total implicit adaptation decreased in the high-variance environment due to changes in error sensitivity, not in forgetting. These observations suggest a model in which the implicit system becomes more sensitive to errors when they occur in a consistent direction. Thus, residual errors in motor adaptation are at least in part caused by an implicit learning system that modulates its error sensitivity in response to the consistency of past errors.

The visual geometry of a tool modulates generalization during adaptation.

Knowledge about a tool's dynamics can be acquired from the visual configuration of the tool and through physical interaction. Here, we examine how visual information affects the generalization of dynamic learning during tool use. Subjects rotated a virtual hammer-like object while we varied the object dynamics separately for two rotational directions. This allowed us to quantify the coupling of adaptation between the directions, that is, how adaptation transferred from one direction to the other. Two groups experienced the same dynamics of the object. For one group, the object's visual configuration was displayed, while for the other, the visual display was uninformative as to the dynamics. We fit a range of context-dependent state-space models to the data, comparing different forms of coupling. We found that when the object's visual configuration was explicitly provided, there was substantial coupling, such that 31% of learning in one direction transferred to the other. In contrast, when the visual configuration was ambiguous, despite experiencing the same dynamics, the coupling was reduced to 12%. Our results suggest that generalization of dynamic learning of a tool relies, not only on its dynamic behaviour, but also on the visual configuration with which the dynamics is associated.

Imagery of movements immediately following performance allows learning of motor skills that interfere.

Motor imagery, that is the mental rehearsal of a motor skill, can lead to improvements when performing the same skill. Here we show a powerful and complementary role, in which motor imagery of different movements after actually performing a skill allows learning that is not possible without imagery. We leverage a well-studied motor learning task in which subjects reach in the presence of a dynamic (force-field) perturbation. When two opposing perturbations are presented alternately for the same physical movement, there is substantial interference, preventing any learning. However, when the same physical movement is associated with follow-through movements that differ for each perturbation, both skills can be learned. Here we show that when subjects perform the skill and only imagine the follow-through, substantial learning occurs. In contrast, without such motor imagery there was no learning. Therefore, motor imagery can have a profound effect on skill acquisition even when the imagery is not of the skill itself. Our results suggest that motor imagery may evoke different neural states for the same physical state, thereby enhancing learning.

Motor Planning, Not Execution, Separates Motor Memories.

Recent theories of limb control emphasize motor cortex as a dynamical system, with planning setting the initial neural state, and execution arising from the self-limiting evolution of the intrinsic neural dynamics. Therefore, movements that share an initial trajectory but then diverge might have different neural states during the execution of the identical initial trajectories. We hypothesized that motor adaptation maps neural states to changes in motor command. This predicts that two opposing perturbations, which interfere when experienced over the same movement, could be learned if each is associated with a different plan even if not executed. We show that planning, but not executing, different follow-through movements allow opposing perturbations to be learned simultaneously over the same movement. However, no learning occurs if different follow throughs are executed, but not planned prior to movement initiation. Our results suggest neural, rather than physical states, are the critical factor associated with motor adaptation.