Reuven Katz

Experimental method for measuring simplicity in mechanical design

ABSTRACT Mechanical designers strive to create a simple design that will reduce products cost and increase its reliability. This paper presents an experimental method for measuring the simplicity of a mechanical design. It also presents a quantitative metric for evaluating the simplicity of product design based on quantifiable design parameters. The simplicity metric was developed by using experts’ evaluations of functionally equivalent products. The dominant parameter for predicting an experts simplicity rating is the number of features required to fully define the examined computer-aided design (CAD) model. This parameter is highly correlated with two additional measured design parameters: number of parts and number of interfaces between parts.

Distinct electroencephalographic responses to disturbances and distractors during continuous reaching movements

Discrepancies between actual and appropriate motor commands, dubbed low-level errors, have been shown to elicit a P300 like component. P300 has been studied extensively in cognitive tasks using, in particular, the three-stimulus oddball paradigm. This paradigm revealed two sub-components, known as P3a and P3b, whose relative contributions depend on saliency and task-relevance, respectively. However, the existence and roles of these sub-components in response to low-level errors are poorly understood. Here we investigated responses to low level errors generated by disturbances - including target and cursor jumps, versus responses to distractors, i.e., environmental changes that are irrelevant to the reaching task. Additionally, we examined the response to matching cursor and target jumps (dual jumps), which generate estimation errors, and are thus considered task relevant disturbances, but do not generate low level errors. We found that a significant P3a-like component is evoked by both disturbances and distractors, whereas the P3b-like component is significantly stronger in response to disturbances than distractors. The P3b-like component appears also in response to dual jumps, even though there are no low level errors. We conclude that disturbances and distractors elicit distinct responses, and that the P3b-like component reflects estimation errors rather than low-level errors.