Palm board and verbal estimates of slant reflect the same perceptual representation
Atten Percept Psychophys (2016) 78:663–673
DOI 10.3758/s13414-015-1029-8
Palm board and verbal estimates of slant reflect
the same perceptual representation
Dennis M. Shaffer 1 & Ally Taylor 1 & Eric McManama 1 &
Allyson Thomas 1 & Echoe Smith 1 & Phil Graves 1
Published online: 24 November 2015
# The Psychonomic Society, Inc. 2015
Abstract People verbally overestimate the orientation of
slanted surfaces, but accurately estimate or underestimate
slanted surfaces using a palm board. We demonstrate a fundamental issue that explains why the two different values typically given for palm board and verbal/visual matching estimates express similar perceptual representations of slanted
surfaces. The fundamental problem in studies measuring palm
board and verbal estimates is that the Bmeasure^—either (1)
reproducing a verbally given angle or the orientation of a
slanted surface with an unseen hand or (2) verbally or visually
estimating a visually perceived surface—has always been
confounded with the Bsurface^—either using (1) a palm board
or (2) a hill or ramp. Although reproduction has exclusively
been used with palm boards in these studies, at the same time
verbal estimation or visual matching has exclusively been
used with hills/ramps. In three experiments, we showed that
verbally estimating palm board orientations produces overestimates by a factor of 1.5, whereas reproducing the orientation
of the surface of a ramp to verbally given angles produces
gains of ~0.6. These values are similar to those seen for verbal
overestimates of slanted surfaces, and to palm board gains for
near surfaces and the relative palm-board-to-verbal gains for
outdoor hills, respectively. Eliminating this confound eliminated the difference previously seen across surfaces. We discuss how and why different measures should produce different results if we overestimate slant in general and perceptually
represent slant in the same way, both haptically and visually.
* Dennis M. Shaffer
1
Department of Psychology, Ohio State University–Mansfield, 1760
University Drive, Mansfield, OH, USA
Keywords Slant perception . Spatial orientation . Pitch
For the last two decades, people have verbally overestimated
the slant of visually perceived geographical, virtual, and manmade hills by between 5° and 25° (Bhalla & Proffitt, 1999;
Creem-Regehr, Gooch, Sahm, & Thompson, 2004; Durgin,
Hajnal, Li, Tonge, & Stigliani, 2010; Durgin & Li, 2011;
Hajnal, Abdul-Malak, & Durgin, 2011; Li & Durgin, 2010;
Proffitt, Bhalla, Gossweiler, & Midgett, 1995; Shaffer & Flint,
2011; Shaffer, McManama, Swank, Williams, & Durgin,
2014; Stigliani, Li, & Durgin, 2013). Two additional measures
have also been used in several of these studies. The first of
these two is visual matching (Bhalla & Proffitt, 1999; CreemRegehr et al., 2004; Proffitt et al., 1995; Witt & Proffitt, 2007).
In the visual matching task, participants manually adjust a pieshaped segment of a disk, which consists of an adjustable
section representing the cross-section of the inclination of
the hill, to be equivalent to the slant of the hill. People overestimate the slant of hills using visual matching, and they do
so by close to the same extent that they do when making
verbal estimates of the hill (Proffitt et al., 1995). The second
of these measures is haptic matching. In these studies, haptic
matching estimates are often made using a palm board—people rest their hand on the palm board and rotate it up from
horizontal to match the palm board to their visual perception
of the slope of the hill (e.g., Bhalla & Proffitt, 1999; Proffitt
et al., 1995; Taylor-Covill & Eves, 2013). Some of these studies have also used a free-hand measure for their haptic
matching task (Bridgeman & Hoover, 2008; Durgin, Hajnal,
et al., 2010, Durgin, Li, et al., 2010; Li & Durgin, 2011;
Shaffer et al., 2014). This consists of holding one’s unseen
hand and/or forearm parallel to the slope. Whereas verbal,
visual matching, and free-hand estimates of visually perceived
hills suggest that hills appear much steeper than their physical
664
inclination, palm board estimates have proven to be more accurate in matching the true inclinations of hills. The reason
given for the apparent difference between palm board and verbal and visual matching estimates has been a dissociation in the
visual pathways in the brain that inform visual awareness and
visually guided action (Bhalla & Proffitt, 1999; Creem &
Proffitt, 1998; Proffitt et al., 1995; Witt & Proffitt, 2007). The
idea is that palm board estimates provide a more accurate representation of the physical world and are a reflection of the
dorsal action stream of the brain that allows us to traverse hills
without a problem, whereas verbal and visual matching estimates are reflections of the ventral conscious perception stream
of the brain, which is more susceptible to illusions and biases.
The ideas that palm boards are accurate, are action measures, are reliable, and do not overestimate the true inclination
of slanted surfaces has recently been brought into question by
work highlighting wrist flexion and anchoring issues with
palm boards (Durgin, Hajnal et al., 2010, Durgin, Li et al.,
2010; Shaffer, McManama, & Durgin, 2015; Shaffer et al.,
2014). For instance, it has been shown that palm boards provide biased and insensitive measures that are a reflection of
the lack of wrist flexion when using them, and not measures
that tap into a distinct visual representation of slant (Durgin,
Hajnal et al., 2010, Durgin, Li et al., 2010). Moreover, the
standard procedure calls for setting the palm board at waist
level as its initial starting position. We have shown that when
palm boards are adjusted from horizontal they give much lower hill matches (by 15° to 30°) than when they are adjusted to
start from vertical, for both near and far surfaces and for ramps
and geographical hills alike (Shaffer, McManama et al., 2015;
Shaffer, Taylor, Thomas, Graves, Smith, & McManama,
2015). This contradicts the idea that palm boards reflect a
more accurate action system for interacting with the physical
world that is independent of a conscious perception system.
Notwithstanding this evidence, recently other measures—
matching extent and remote haptic perception—have been
supportive of a single underlying representation that is exaggerated to the same extent as are verbal and visual matching
estimates (Li & Durgin, 2010; Shaffer & McManama, 2015).
Li and Durgin (2010) had one group of people compare the
relative length of a frontal extent to the length of an extent that
was placed up a virtual hill (tilted back in depth). This provided, by trigonometry, an implicit estimate of perceived slant.
The other group gave verbal estimates of the same hill. The
implicit slant measure gave the same results as the verbal
estimates—both showed the same exaggeration of all slants.
Shaffer and McManama (2015) used a remote haptic device in
which the participant held one end of a wood (...truncated)
