Latency to categorize disoriented alphanumeric characters as letters or digits.

Abstract

Previous research has shown that if Ss must decide quickly whether a disoriented alphanumeric character is normal or backward, they "mentally rotate" an internal representation of the character to its normal upright position before making the decision. The present study with 12 undergraduates showed that latency to decide whether a disoriented character was a letter or a digit was barely affected by angular orientation, implying that mental rotation is not necessary for this task. However, latency was significantly shorter for normal than for backward characters, regardless of angular orientation, suggesting that the difference between normal and backward characters does register prior to mental rotation. (7 ref)

Recognition of forms rotated in depth

“…Unfortunately, this alternative explanation has at least two weak points: First, it does not provide a reasonable basis to account for the cases in which no increase in reaction time was observed regardless of increasing angular disparity around a z-axis (Corballis & Nagourney, 1978;Eley, 1982;White, 1980). The results in the Identity and Combination conditions of the present study contradict their explanation more directly because rotation around a yaxis was used just as in their study.…”

“…It was once suggested that recognition of a form is always accomplished by mentally rotating its viewercentered representation when the viewpoint has changed. Unfortunately, this alternative is unable to explain the cases in which mental rotation is not carried out (Corballis & Nagourney, 1978;Corballis, Zbrodoff, Shetzer, & Butler, 1978;Eley, 1982;White, 1980). The information type theory (Takano, 1987(Takano, , 1989 has provided an integrated explanation for both the case where form recognition is affected by viewpoint change and the case where it is not.…”

On the process of recognizing inverted words: Does it rely only on orientation-invariant cues?

“…It is thus alternatively possible that rectification is applied in a piecemeal fashion, possibly on a letter-by-letter basis. On the arguable assumption that letter recognition does not require any normalization (see, e.g., Corballis & Nagourney, 1978;but see Jordan & Huntsman, 1995)-at least not for cursory identification (Corballis, 1988)-one might doubt that mental rotation is needed for a letter-by-letter process. In any event, any hypothesis that posits such a piecemeal process, regardless of whether or not its first stage involves letter normalization, would require a letter reordering process at its second stage.…”

“…However, several arguments suggest that this possibility is not quite viable. First, the identification of single letters or letterlike stimuli might be affected little by planar orientation (Corballis & Nagourney, 1978;Corballis, Zbrodoff, Shetzer & Butler, 1978;Eley, 1982;Koriat & Norman, 1989;White, 1980; but see Jolicoeur & Landau, 1984;Jordan & Huntsman, 1995;McMullen & Jolicoeur, 1990). Were that indeed the case, it would be wasteful if the product of single letter identification was not used in the process of reading inverted texts.…”

Can the Comparisons of Feature Locations Explain the Difficulty in Discriminating Mirror-Reflected Pairs of Geometrical Figures from Disoriented Identical Pairs?

“…Cohen and Kubovy [13] set a limiting rate of 1 ms/°, below which the process of mental rotation was considered not to be involved in figure discriminations. For example, studies requiring participants to name disoriented alphanumeric characters [14], the letter/digit distinction of disoriented alphanumeric characters [15], the naming of disoriented drawings of natural objects [16] and top/bottom decisions of the location of a dot mark near the top or bottom of disoriented objects [17] resulted in mental rotation rates of less than 1 ms/°. In contrast, the same/mirror-reflected decisions about 3D objects [1,6,18], the same/different matching of the surfaces of cubes [19] and the same/mutant decision about randomly generated polygons [11] are examples of studies in which the rates of mental rotation were greater than 20 ms/°.…”