Comment: Frontal lobes, executive dysfunction, gait and the fallacy of Pseudo-transitivity

Freezing of gait (FOG) in patients with Parkinson disease (PD) is particularly vexing as it causes disability and yet is often treatment refractory. Further, gait disorders do not track with volitional motor control. For example, gait often worsens with high-frequency deep brain stimulation while upper extremity function improves. Frontal lobes lesions can produce Bruns ataxia, an apraxia restricted to gait.

Frontal lobe involvement in gait is further substantiated for PD by the findings of Kostić et al. showing relatively greater gray matter loss in the dorsolateral prefrontal cortex (along with other cortical areas). Such studies are difficult because of diagnostic imprecision for FOG, variability of occurrence, and effects of the visual environment, but their observations, as opposed to their inferences, are otherwise robust and important.

The authors infer some causal relation between FOG and executive dysfunction based on commonality of frontal lobe pathology. While possibly true, the argument is an example of the fallacy of pseudotransitivity, where “executive dysfunction implies frontal lobe pathology and FOG implies frontal lobe pathology; therefore executive dysfunction implies FOG.” Such reasoning is acceptable for hypothesis generation but it is insufficient for formulating conclusions.

The problem with this line of reasoning is a resulting presumption of executive dysfunction as causal to FOG. This leads to the fallacy of limited alternatives, where one presents a limited range of explanations (or brain regions) and proceeds to ignore all but a few. The result is undue support for the remaining possibilities. In other words, “To the man who only has a hammer, everything he encounters begins to look like a nail” (attributed to Abraham Maslow). As evidence, the authors imply that frontal executive dysfunction leads to pedunculopontine nucleus dysfunction and FOG, although alternative explanations for pedunculopontine nucleus involvement exist. This study is an important first step; we await further investigations to see where it may lead.

• Copyright © 2012 by AAN Enterprises, Inc.