Abstract
Previous studies suggested that initial-contact error rates were higher for touch input devices (resistive and infrared) than for various hardware keys. A follow-up study was conducted to evaluate the effects of chemical/biological/radiation (CBR) protective gloves on performance. Error data from this and the preceding study in this series were somewhat misleading due to differences in the input sampling strategies between hardware keys and touch panels. The initially high error rates for the resistive panel (21.6% using flight gloves and a 26.1% using CBR gloves) were considered unrepresentative of a true operating environment because nonkey contacts would be ignored by the system. Revision of the input-sensing logic produced 100% accuracy for the resistive panel in the replication study. However, multiple contacts were required in 5% (median of 4.1%) of the trials. Overall, 4% required only a second contact for correct activation of the designated key while 3- and 4-contact activations amounted to 0.6% of all trials. Conclusions drawn from the research were that CBR gloves do not significantly degrade performance over standard flight gloves or, for that matter, over the bare hand (errors or response times) and that the occurrence of multiple touches is small and virtually disappears with experience and adequate feedback, errors usually occurring in early trials and in no-feedback conditions. Thus, systems using gloved operators are unlikely to experience serious performance degradation if adequate key sizes for bare-handed activation are used and if timely and relevant feedback is provided to the operator.
Get full access to this article
View all access options for this article.