Limits...
The Subjective Visual Vertical and the Subjective Haptic Vertical Access Different Gravity Estimates.

Fraser LE, Makooie B, Harris LR - PLoS ONE (2015)

Bottom Line: Experiment 1 compared SVV and SHV across three levels of whole-body tilt and found that SVV showed an A-effect at larger tilts while SHV was accurate.Experiment 2 found that tilting either the head or the trunk independently produced an A-effect in SVV while SHV remained accurate when the head was tilted on an upright body but showed an A-effect when the body was tilted below an upright head.Overall our results suggest that SVV and SHV access distinct underlying gravity percepts based primarily on head and body position information respectively, consistent with a model proposed by Clemens and colleagues.

View Article: PubMed Central - PubMed

Affiliation: Center for Vision Research, York University, Toronto, Ontario, Canada.

ABSTRACT
The subjective visual vertical (SVV) and the subjective haptic vertical (SHV) both claim to probe the underlying perception of gravity. However, when the body is roll tilted these two measures evoke different patterns of errors with SVV generally becoming biased towards the body (A-effect, named for its discoverer, Hermann Rudolph Aubert) and SHV remaining accurate or becoming biased away from the body (E-effect, short for Entgegengesetzt-effect, meaning "opposite", i.e., opposite to the A-effect). We compared the two methods in a series of five experiments and provide evidence that the two measures access two different but related estimates of gravitational vertical. Experiment 1 compared SVV and SHV across three levels of whole-body tilt and found that SVV showed an A-effect at larger tilts while SHV was accurate. Experiment 2 found that tilting either the head or the trunk independently produced an A-effect in SVV while SHV remained accurate when the head was tilted on an upright body but showed an A-effect when the body was tilted below an upright head. Experiment 3 repeated these head/body configurations in the presence of vestibular noise induced by using disruptive galvanic vestibular stimulation (dGVS). dGVS abolished both SVV and SHV A-effects while evoking a massive E-effect in the SHV head tilt condition. Experiments 4 and 5 show that SVV and SHV do not combine in an optimally statistical fashion, but when vibration is applied to the dorsal neck muscles, integration becomes optimal. Overall our results suggest that SVV and SHV access distinct underlying gravity percepts based primarily on head and body position information respectively, consistent with a model proposed by Clemens and colleagues.

Show MeSH

Related in: MedlinePlus

A simplified version of the model presented in Clemens et al [3], with several additions.The original model posits two final position estimates used to determine the direction of gravity, one based on the head’s position (with indirect input from body-based sensors, converted via neck proprioception into head-centric coordinates) and the other on the body (with indirect input from the head). The authors suggest SVV uses the head-based gravity estimate; here we argue that SHV accesses the body-based estimate.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4696803&req=5

pone.0145528.g001: A simplified version of the model presented in Clemens et al [3], with several additions.The original model posits two final position estimates used to determine the direction of gravity, one based on the head’s position (with indirect input from body-based sensors, converted via neck proprioception into head-centric coordinates) and the other on the body (with indirect input from the head). The authors suggest SVV uses the head-based gravity estimate; here we argue that SHV accesses the body-based estimate.

Mentions: When only the head is tilted (with the body upright) SHV is accurate but biases in SVV are still found [11] which suggests that head tilt may drive SVV errors, but not SHV errors. SHV errors, in contrast, may be driven by the tilted position of the torso and limbs; this is supported by Bauermeister et al. [20] who found that SHV shifts position depending on which hand is used to feel the test probe. Schuler and colleagues [2] have suggested that different errors in SVV and SHV may reflect modality-specific errors, such as the hand’s position being misestimated if the tilt and torsion of the joints of the arm are misperceived. According to this theory, errors in SVV and SHV may have more to do with errors in comparing the test probe to an internal estimate of gravity than to errors in the actual estimate itself. In other words, response errors could reflect misperception of the probe’s orientation as sensed by a specific modality, not the orientation of gravity. Such modality specific errors are ostensibly head based for vision and limb/torso based for haptic measures. An alternative theory by Clemens and colleagues [3] suggests that there may be multiple, simultaneous internal estimates of gravity, held in different reference frames (head based or body based) that may not necessarily be congruent and which may be differentially accessed by different verticality judgment tasks (see Fig 1). If this is the case, and if SVV and SHV access different estimates within this system, then SVV and SHV should be able to combine in a statistically optimal fashion into a bimodal estimate of gravity. That is, if the noise in SVV and SHV estimates are uncorrelated then a bimodal estimate of gravitational vertical should be the average of visual and haptic responses, weighted by the respective reliabilities of the two measures, with an increased precision compared to unimodal responses [23,24]. In experiments 4 and 5 we test whether a bimodal probe of subjective vertical is consistent with these predictions of optimal cue combination.


The Subjective Visual Vertical and the Subjective Haptic Vertical Access Different Gravity Estimates.

Fraser LE, Makooie B, Harris LR - PLoS ONE (2015)

A simplified version of the model presented in Clemens et al [3], with several additions.The original model posits two final position estimates used to determine the direction of gravity, one based on the head’s position (with indirect input from body-based sensors, converted via neck proprioception into head-centric coordinates) and the other on the body (with indirect input from the head). The authors suggest SVV uses the head-based gravity estimate; here we argue that SHV accesses the body-based estimate.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4696803&req=5

pone.0145528.g001: A simplified version of the model presented in Clemens et al [3], with several additions.The original model posits two final position estimates used to determine the direction of gravity, one based on the head’s position (with indirect input from body-based sensors, converted via neck proprioception into head-centric coordinates) and the other on the body (with indirect input from the head). The authors suggest SVV uses the head-based gravity estimate; here we argue that SHV accesses the body-based estimate.
Mentions: When only the head is tilted (with the body upright) SHV is accurate but biases in SVV are still found [11] which suggests that head tilt may drive SVV errors, but not SHV errors. SHV errors, in contrast, may be driven by the tilted position of the torso and limbs; this is supported by Bauermeister et al. [20] who found that SHV shifts position depending on which hand is used to feel the test probe. Schuler and colleagues [2] have suggested that different errors in SVV and SHV may reflect modality-specific errors, such as the hand’s position being misestimated if the tilt and torsion of the joints of the arm are misperceived. According to this theory, errors in SVV and SHV may have more to do with errors in comparing the test probe to an internal estimate of gravity than to errors in the actual estimate itself. In other words, response errors could reflect misperception of the probe’s orientation as sensed by a specific modality, not the orientation of gravity. Such modality specific errors are ostensibly head based for vision and limb/torso based for haptic measures. An alternative theory by Clemens and colleagues [3] suggests that there may be multiple, simultaneous internal estimates of gravity, held in different reference frames (head based or body based) that may not necessarily be congruent and which may be differentially accessed by different verticality judgment tasks (see Fig 1). If this is the case, and if SVV and SHV access different estimates within this system, then SVV and SHV should be able to combine in a statistically optimal fashion into a bimodal estimate of gravity. That is, if the noise in SVV and SHV estimates are uncorrelated then a bimodal estimate of gravitational vertical should be the average of visual and haptic responses, weighted by the respective reliabilities of the two measures, with an increased precision compared to unimodal responses [23,24]. In experiments 4 and 5 we test whether a bimodal probe of subjective vertical is consistent with these predictions of optimal cue combination.

Bottom Line: Experiment 1 compared SVV and SHV across three levels of whole-body tilt and found that SVV showed an A-effect at larger tilts while SHV was accurate.Experiment 2 found that tilting either the head or the trunk independently produced an A-effect in SVV while SHV remained accurate when the head was tilted on an upright body but showed an A-effect when the body was tilted below an upright head.Overall our results suggest that SVV and SHV access distinct underlying gravity percepts based primarily on head and body position information respectively, consistent with a model proposed by Clemens and colleagues.

View Article: PubMed Central - PubMed

Affiliation: Center for Vision Research, York University, Toronto, Ontario, Canada.

ABSTRACT
The subjective visual vertical (SVV) and the subjective haptic vertical (SHV) both claim to probe the underlying perception of gravity. However, when the body is roll tilted these two measures evoke different patterns of errors with SVV generally becoming biased towards the body (A-effect, named for its discoverer, Hermann Rudolph Aubert) and SHV remaining accurate or becoming biased away from the body (E-effect, short for Entgegengesetzt-effect, meaning "opposite", i.e., opposite to the A-effect). We compared the two methods in a series of five experiments and provide evidence that the two measures access two different but related estimates of gravitational vertical. Experiment 1 compared SVV and SHV across three levels of whole-body tilt and found that SVV showed an A-effect at larger tilts while SHV was accurate. Experiment 2 found that tilting either the head or the trunk independently produced an A-effect in SVV while SHV remained accurate when the head was tilted on an upright body but showed an A-effect when the body was tilted below an upright head. Experiment 3 repeated these head/body configurations in the presence of vestibular noise induced by using disruptive galvanic vestibular stimulation (dGVS). dGVS abolished both SVV and SHV A-effects while evoking a massive E-effect in the SHV head tilt condition. Experiments 4 and 5 show that SVV and SHV do not combine in an optimally statistical fashion, but when vibration is applied to the dorsal neck muscles, integration becomes optimal. Overall our results suggest that SVV and SHV access distinct underlying gravity percepts based primarily on head and body position information respectively, consistent with a model proposed by Clemens and colleagues.

Show MeSH
Related in: MedlinePlus