Pitching people with an inversion table: Estimates of body orientation are tipped as much as those of visual surfaces

Atten Percept Psychophys (2016) 78:700–706 DOI 10.3758/s13414-015-1019-x Pitching people with an inversion table: Estimates of body orientation are tipped as much as those of visual surfaces Dennis M. Shaffer 1 & Ally Taylor 1 & Allyson Thomas 1 & Phil Graves 1 & Echoe Smith 1 & Eric McManama 1 Published online: 10 November 2015 # The Psychonomic Society, Inc. 2015 Abstract In the current work we investigate people’s perception of their own body tilt in the pitch direction. In Experiment 1, we tilted people backward at 1 of 5 different randomly assigned angles using an inversion table. People significantly overestimated the angle at which they were tilted backward at angles from 8° to 45°. The slope of the plotted average overestimates had a gain of 1.46, fitting nicely with previously reported gains of verbal overestimates of visually perceived slant of natural outdoor geographically oriented slopes as well as man-made wooden slopes within and outside of reach in the laboratory. In Experiment 2, we showed participants a 45° line and asked them to indicate when they were positioned at that orientation. Participants again significantly overestimated the angle at which they were tilted backward. This extends work showing that a scale-expanded theory of visual space is multisensory, results in equivalent estimates for both verbal and nonverbal/nonnumeric methods, and can now be expanded to include the perceived orientation of one’s own body. Keywords Slant perception . Spatial orientation . Pitch For the last two decades, a wealth of evidence shows that people overestimate the slant of both geographical and manmade slopes by between 5° and 25° (Bhalla & Proffitt, 1999; Bridgeman & Hoover, 2008; Creem & Proffitt, 1998; CreemRegehr, Gooch, Sahm, & Thompson, 2004; Durgin & Li, 2011; Durgin, Li, & Hajnal, 2010; Hajnal, Abdul-Malak, & * Dennis M. Shaffer 1 Department of Psychology, The Ohio State University, Mansfield, 1760 University Drive, Mansfield, OH 44906, USA Durgin, 2011; Li & Durgin, 2010; Proffitt, Bhalla, Gossweiler, & Midgett, 1995; Proffitt, Creem & Zosh, 2001; Shaffer & Flint, 2011; Shaffer & McManama, 2015; Stefanucci, Proffitt, Clore, & Parekh, 2008; Witt & Proffitt, 2007). Much less work has been performed on people’s perception of their own body orientation in the pitch dimension, and the results of some of this work are difficult to interpret. For instance, Cohen and Larson (1974) had participants adjust the pitch of their own body every 15° from a supine position to a prone position while restrained in a motorized hospital bed. They found systematic errors of underestimation of body tilt. For instance, when asked to place themselves at 15° backward from a vertical position, they placed themselves at 29° backward, and when asked to place themselves at 15° forward, they placed themselves at 23° forward. These errors were consistent but smaller as they moved in either direction in 15° increments from 15° to 60°, at which point there was almost no error. Of the studies to investigate pitch perception, this seems to be the only one where people underestimate pitch. We feel there are at least two reasons for this. First, participants were moved backward in 15° increments until they were prone, and then forward in 15° increments until they were in a supine position. They did this back and forth a total of eight times (four backward, four forward). Carryover effects from each previous estimate likely affected their subsequent estimate. Second, they were giving estimates of, say, 15° backward when they were either erect (straight up and down) or oriented at 30° backward (depending on whether they were in the forward or backward sequence). Separate analyses were not reported for forward and backward sequences, so it is not known whether there were anchoring biases that could strongly affect their estimates and whether these were symmetrical or not (Shaffer, McManama, Swank, Williams, & Durgin, 2014; Shaffer, McManama, & Durgin, in press). Atten Percept Psychophys (2016) 78:700–706 More recently, Ito and Gresty (1997) assessed postural orientation in the pitch dimension while participants were in seated and standing positions. Participants were instructed to estimate when they felt like they were tilted 45° forward and backward and 90° forward and backward. When their head, trunk, and legs were in alignment (BProtocol 5^) and they estimated they were at 45° backward, they were actually positioned at 32° backward (from erect), and when they estimated they were at 45° forward they were actually positioned at 37° forward. This provides evidence that people overestimate how much they are tilted in the pitch dimension. However, participants either tilted themselves or were tilted by the experimenter 45° forward prior to estimating when they were tilted backward at 45°, so it is difficult to say whether or by how much participants were affected by this, especially in the seated condition where their estimates of 45° backward averaged 4° backward and their estimates of 90° backward averaged 47° backward. Jewell (1998) was able to control for some of the factors making it difficult to interpret the aforementioned work by asking participants to estimate only one angle. He performed a series of experiments tilting people backward in chairs, Bhand trucks,^ or a gyroscope with eyes open and eyes closed. Participants were tilted from vertical (Study 1, Experiments 1 & 2 and Study 2, Experiment 1) and asked to indicate when they were tilted backward at a 45° angle. When participants were tilted backward while seated in chairs or while standing in hand trucks estimates were significantly earlier than 45°. When participants stood in the gyroscope and were tilted backward, they estimated they were tilted at 45° when they were only tilted at 32.94° (eyes open), similar to Ito and Gresty (1997), and 28.74° (eyes closed). The current work had four purposes. First, we wanted to extend previous work, by performing a systematic investigation of people’s perception of their own body tilt in the pitch dimension. Therefore, we tilted people backward at five different randomly assigned angles (8°, 16°, 24°, 32°, and 40°) and recorded their estimates of the angles at which they perceived they were tilted. We did this to test whether there is a pattern of overestimation across a wide range of angles. Second, we wanted to tilt people backward from straight up and down until they perceived they were tilted backward at 45° and compare these results with those of Jewell (1998). Third, we wanted to have people match their body to a nonverbal diagram of a 45° line in order to test whether a nonverbal and nonnumeric method would yield equivalent results to verbal estimations. Finally, we sought to investigate whether the pattern of results we would find for proximal slant or the perceived orientation of one’s own body at the five different angles is similar to the pattern seen for the verbal (...truncated)