We used 3 psychophysics methods to determine perceptible changes in seat he
ight, seat pan angle, and backrest angle using an experimental chair. In th
e method of adjustment, the chosen chair settings were affected by the init
ial setting. For example, a high initial setting of the seat height led to
a high selected setting and a low setting led to a low value. The differenc
e between settings was referred to as not noticeable difference (NND). The
method of limits was used to determine acceptable chair settings using verb
al limits such as "too high" and "too low." Using the method of constant st
imuli, just noticeable differences (JNDs) were determined for chair height
(1.5 cm), seat pan angle (1.2 degrees) and backrest angle (1.7 degrees). Th
e corresponding values for NNDs and verbal limits were about twice as large
: chair height (2.5 cm), seat pan angle (4 degrees) and backrest angle (3 d
egrees). NNDs and verbal limits are unobtrusive measures that are considere
d more valid than JNDs, which exaggerate the need for adjustability. The re
sults have practical implications for the design of office chairs.