Patricia A. McMullen

Effects of Orientation on the Identification of Rotated Objects Depend on the Level of Identity

Matching names and rotated line drawings of objects showed effects of object orientation that depended on name level. Large effects, in the same range as object naming, were found for rotations between 0 degrees and 120 degrees from upright with subordinate names (e.g., collie), whereas nonsignificant effects were found with superordinate (e.g., animal) and basic names (e.g., dog). These results support image normalization, after contact with orientation-invariant representations, that provide basic-level identity. They consequently fail to support theories of object recognition in which rotated object images are normalized to the upright position before contact with long-term object representations.

The spatial frame of reference in object naming and discrimination of left-right reflections

The effects of stimulus rotation and observers head-tilt position on various pattern-recognition tasks were investigated to compare the external directions most closely aligned with the spatial frame of reference. Specifically, the effects of these factors on the time to name objects were compared with their effects on the time to discriminate left-facing from right-facing lateral views of these objects, as well as with their effects on the time to discriminate normal from mirrorimaged alphanumeric characters. The naming task relied upon a reference frame more closely aligned with retinal directions than with environmental directions. In contrast,both discrimination tasks relied upon a frame aligned more closely with environmental directions. Overall, the nature of the task exerts a greater influence on the directions with which the frame of reference is aligned than do the stimulus attributes.

Reference frame and effects of orientation on finding the tops of rotated objects

Effects of stimulus orientation across trial blocks and the spatial reference frame were investigated with a task in which Ss, with their head upright or tilted, judged a dot to be near the top or the bottom of rotated line drawings of objects. Objects used in this task were also named. Response times from the first block of trials increased linearly for objects rotated from 0 degrees to 120 degrees from the upright. Across blocks, orientation effects diminished for naming but remained the same for top-bottom discriminations. Practice with top-bottom discriminations diminished orientation effects when the same objects were subsequently named. The spatial reference frame for top-bottom discriminations was midway between retinal and environmental coordinates. Specifying the location of object features is of greater importance for top-bottom discriminations than for naming and underlies orientation effects in these tasks.

Subordinate-level categorization relies on high spatial frequencies to a greater degree than basic-level categorization

In two experiments, category verification of images of common objects at subordinate, basic, and superordinate levels was performed after low-pass spatial filtering, high-pass spatial filtering, 50% phase randomization, or no image manipulation. Both experiments demonstrated the same pattern of results: Low-pass filtering selectively impaired subordinate-level category verification, while having little to no effect on basic-level category verification. Subordinate categorization consequently relies to a greater degree on high spatial frequencies of images. This vulnerability of subordinate-level processing was specific to a lack of high spatial frequency information, as opposed to other visual information, since neither high-pass filtering nor the addition of phase noise produced a comparable reduction in performance. These results are consistent with the notion that object recognition at basic levels relies on the general shapes of objects, whereas recognition at subordinate levels relies on finer visual details.

Orientation-invariant transfer of training in the identification of rotated natural objects

The effects of stimulus orientation on naming were examined in two experiments in which subjects identified line drawings of natural objects following practice with the objects at the same or different orientations. Half the rotated objects were viewed in the orientation that matched the earlier presentations, and half were viewed at an orientation that mismatched the earlier presentations. Systematic effects of orientation on naming time were found during the early presentations. These effects were reduced during later presentations, and the size of this reduction did not depend on the orientation in which the object had been seen originally. The results are consistent with a dual-systems model of object identification in which initially large effects of disorientation are the result of a normalization process such as mental rotation, and in which attenuation of the effects is due to a shift from the normalization system to a feature/part-based

Face Recognition Is Affected by Similarity in Spatial Frequency Range to a Greater Degree Than Within-Category Object Recognition.

Previous studies have suggested that face identification is more sensitive to variations in spatial frequency content than object recognition, but none have compared how sensitive the 2 processes are to variations in spatial frequency overlap (SFO). The authors tested face and object matching accuracy under varying SFO conditions. Their results showed that object recognition was more robust to SFO variations than face recognition and that the vulnerability of faces was not due to reliance on configural processing. They suggest that variations in sensitivity to SFO help explain the vulnerability of face recognition to changes in image format and the lack of a middle-frequency advantage in object recognition.

Perceptual biases in processing facial identity and emotion

Normal observers demonstrate a bias to process the left sides of faces during perceptual judgments about identity or emotion. This effect suggests a right cerebral hemisphere processing bias. To test the role of the right hemisphere and the involvement of configural processing underlying this effect, young and older control observers and patients with right hemisphere damage completed two chimeric faces tasks (emotion judgment and face identity matching) with both upright and inverted faces. For control observers, the emotion judgment task elicited a strong left-sided perceptual bias that was reduced in young controls and eliminated in older controls by face inversion. Right hemisphere damage reversed the bias, suggesting the right hemisphere was dominant for this task, but that the left hemisphere could be flexibly recruited when right hemisphere mechanisms are not available or dominant. In contrast, face identity judgments were associated most clearly with a vertical bias favouring the uppermost stimuli that was eliminated by face inversion and right hemisphere lesions. The results suggest these tasks involve different neurocognitive mechanisms. The role of the right hemisphere and ventral cortical stream involvement with configural processes in face processing is discussed.

Manipulability and living/non-living category effects on object identification

Object naming studies have generally observed that both normal and brain damaged individuals are faster and more accurate at identifying non-living objects than living objects (). However, a potential confounding variable, manipulability, has been present in past studies that may mediate this effect. Previous studies that have observed a non-living advantage have often used manipulable and non-manipulable exemplars to represent the non-living and living groups, respectively. Under conditions which controlled for object manipulability and familiarity, results demonstrated advantages for the identification of non-manipulable and for living objects.

Norms for two types of manipulability (graspability and functional usage), familiarity, and age of acquisition for 320 photographs of objects

There is increasing interest in the role that manipulability plays in processing objects. To date, Magnié, Besson, Poncet, and Dolisi’s (2003) manipulability ratings, based on the degree to which objects can be uniquely pantomimed, have been the reference point for many studies. However, these ratings do not fully capture some relevant dimensions of manipulability, including whether an object is graspable and the extent to which functional motor associations above and beyond graspability are present. To address this, we collected ratings of these dimensions, in addition to ratings of familiarity and age of acquisition (AoA), for a set of 320 black-and-white photographs of objects. Familiarity and AoA ratings were highly correlated with previously reported ratings of the same dimensions (r =.853, p<.001, and r=.771, p<.001, respectively), validating the present norms. Grasping and functional use ratings, in contrast, were more moderately correlated with Magnié et al.’s pantomime manipulability ratings (r=.507, p<.001). These results were taken as evidence that the new manipulability ratings collected in this research capture distinct aspects of object manipulability. The complete stimuli and norms from this study may be downloaded from http://brm.psychonomic-journals.org/content/supplemental.

ROTATED OBJECT IDENTIFICATION WITH AND WITHOUT ORIENTATION CUES

The time to name two-dimensional line drawings of objects increases linearly for object rotations between 0 degrees and 120 degrees from the upright. Several theories attribute these effects of orientation to finding the top or the top-bottom axis of objects. By this account, prior knowledge of the location of the top or the top-bottom axis of objects should diminish effects of object orientation when they are named. When this hypothesis was tested by cuing the top or the top-bottom axis, no reduction in the effects of orientation on object naming was found. This result is inconsistent with effects of orientation on object naming being due to finding the top or the top-bottom axis. Instead, the top may be found prior to rotational normalization of the object image.