Members of helmet Standards committees frequently need to define limits for test procedures. Such limits relate to test values that indicate the potential for injury and yet it is often difficult for members to know the type and severity of injury that is represented by a given test value. Over the years, criteria have been developed for different body regions and usually these have been derived from a combination of accident and casualty data, and tests on cadavers, cadaver body parts, animals and human volunteers. However, such criteria are often used by the automotive industry as pass/fail values without a clear understanding of human tolerance to injurious forces. This sometimes leads to the mistaken belief that any value below the stated limit implies uninjured and all values above imply a serious or fatal injury. This misconception gives very little freedom to choose values that are different from the often-inappropriate automotive value. This is particularly true for head injury criteria for which values for a helmeted head may be different to those for the unhelmeted head. Many accidents to wearers of helmets, which cover a wide range of activities from horse riding to downhill skiing, result in a closed head injury. This is when the brain is damaged without any skull or external tissue damage. Conversely, head injuries in automotive accidents are much more frequently open head injuries with skull fracture and soft tissue lesions. Other misconceptions arise because of the failure to understand that human response to a given dose or injurious parameter varies across a range of the population. The dose response curve tends to be "S" (sigmoid) shaped such that as the magnitude of the injurious parameter increases so does the percent of the population that sustains an injury of a given severity.