Cornell University Ergonomics Web

DEA3500: Ambient Environment: Thermal Conditions/Health & Performance


Direct effects:

Early factory studies
Vernon (1919) -- British Industrial Fatigue Board. He showed that in hot industries with heavy work there is a lower output in summer than in winter. Vernon also found that accidents increase in munitions factories as temperature falls or decreases from optimum around 20C. Similar results for increased accidents with increased temperature were found for studies in coal mines (Vernon, 1927).



Both hot and cold conditions can impair the performance of a variety of activities. Both warm and cool conditions can influence arousal, and both can decrease this and make one feel drowsy.

Psychomotor performance studies

Wilkinson et al.(1964) - Auditory vigilance task. Subjects' body temperature maintained by a vapor barrier suit. Rectal temperature measured. Subjects had to detect infrequent longer duration tones embedded in a series of tones. Decreased vigilance performance indicates decreased arousal. Results showed that when core temperature was increased slightly to 37.3C, performance increased, but higher temperatures led to speeding of reactions. Performance on a secondary adding task did, however, decrease.

Mackworth (1952) -- looked at errors made by Morse Code operators and found that all subjects had more errors when SET >33C, but that the less skilled operators showed even greater performance decrements.

(SET = 33C == air temperature 33C
RH 50%
low air speed and light clothing)

Wing (1965) -- reviewed studies of temperature and mental task performance. Wing found that there is evidence for a curvilinear relationship between exposure time and exposure temperature and task performance decrements. Wing expressed temperature as the WBGT index (Wet-bulb Globe Thermometer) which is an index preferred by NWSH for hot environments.

WBGT = 0.7 Twb + 0.2 Tg + 0.1 Ta (w = wet bulb, g = globe temperature, a = air temperature)

General agreement that ability to perform mental tasks decrease above 33C air temperature (30CET). 

Hancock (1981) has developed equal-tolerance curves for 3 types of work activities and physiological tolerance.

Temperature and Learning: Classroom Studies
Schoer & Shaffran (1973) -  Lennox Research School, Iowa School constructed to help with development of environmental control systems for schools and 2 identical classrooms tested.

1 classroom - air conditioned and maintained at 22.5C
1 classroom - unconditioned and maintained at 26C
Study took approximately 9 weeks. 19 different tests employed over 3 studies. 6 tests better in air-conditioned classroom (all more complex tasks). None better in unconditioned classrooms.

Pepler (1971) -  studied 3 pairs of schools near Portland, Oregon
3 schools air-conditioned and maintained 24C
3 schools unconditioned and variable up to 29C
Students tested twice per week. Test performance was significantly related to temperature. Generally, results showed that test scores were better with temperatures of 22-23C than >26C.

Laboratory Studies
Pepler & Warner (1968) - Kansas State Univ. climate chamber. 72 undergrads studied self-teaching text under 6 temps. (at 6 different times - rep. measures) in range 16.7C--33.3C at 45% RH.

Students worked faster at high and low temperatures, but also made more errors and reported more effort. C = optimum temperature for learning (= Fanger's comfort temperature).

Wyon (1970) -- Reported on experiments to test language learning in a language laboratory at different temperatures.

Performance was significantly worse at 27C than at 20C, and the effect was more pronounced for less able children. Again, mild heat serves to decrease arousal and high heat to increase this.

Office Studies
1914 - New York State Commission on Ventilation conducted experiments on temperature and behavior.

Hask - Typing with no incentives or knowledge of results at 20C and 24C. Results showed subjects produced less work at 24C than 20C.
Langkilde (1978) - thorough lab. study of office tasks in Copenhagen. Subjects exposed to 5 different temperatures. Over 15 days.
Subjects worked more slowly at 21C than 24C, but fast at 18C than 24C. (At 24C subjects were thermally neutral.) Possible that mild cooling decreases arousal but more severe cooling raises this. For the range of temperatures normally found in offices little effect of temperature or performance may be expected.

Factory Studies
In many factories, workers can be exposed to high radiant temperatures (e.g. 60C in glassworks, steelworks, etc.). Heat exposure limits have been developed to distinguish "safe" and "unsafe" levels.

ASHRAE has set WGBT limits for different workloads for continuous work (8 hour day) and different work-rest rates.

Also, different limits as a function of different air velocities.

COLD STRESS (Refrigerators/cold stores, Arctic/Antarctic Sea, oil rigs/divers etc.

Less of a hazard than heat stress because:

But it leads to hypothermia i.e. core temperature below 35C, which can be fatal.

Shivering is body's reaction to cold, but temperature onset of shivering depends on skin temperature, which in turn depends on fat thickness of subjects.

People with more subcutaneous fat have more insulation against direct conduction losses through skin when cold (i.e. when there is peripheral vasoconstriction). Note when hot, heat loss is via peripheral vasodilation not conduction, that is, there is little difference between thin and fat people.
Fat people start shivering at lower skin temperatures and it takes longer for shivering to start. Also Keating (1969) showed that after 30 minutes in water at 10C, body temperature decreased 2C for thin people but was virtually unchanged for people with 20 mm fat thickness. In part this may be why we tend to increase weight in winter months and decrease in summer.

Cold Acclimatization
Acclimatization to heat is a well-established response to repeated exposure to heat stress. It is not even clear whether general cold acclimatization actually occurs.

Local acclimatization
Good evidence for local adaptation of hands exposed to cold. This occurs because of increased blood flow to hands.

Eskimos have increased blood flow to hands which allows them to maintain dexterity at temperatures below those for Europeans. Also, in cold pressor test (10C), Eskimos felt no pain, only cold, but Europeans felt deep, aching pain. The same effects have been found with Arctic and Northern fishermen. Using hands in cold can affect manual dexterity by decreasing flexibility of digits and decreasing sensitivity. At a critical temperature of fingers at 6C, there is a dramatic decrease in sensitivity as fingers become numb. Air speed is an important contributor to cooling of skin temperature.

Wind Chill Index

WCI = (10 (square root of v) + 10.45 - v)(33 - Ta)
where v = air velocity (m/sec) and Ta = air temperature
WCI effectively expresses rate of cooling of skin.
Gloves - can substantially protect fingers from wind chilling and cold, although dexterity may be decreased.

Go to the next lecture