Goal-directed aiming under restricted viewing conditions with confirmatory sensory feedback
Introduction
For more than a half century, under a wide variety of circumstances, research has established that goal-directed movements are subject to a speed-accuracy tradeoff. That is, when the target of movement is made smaller, or placed farther away, it requires more time to acquire. Fitts (1954), in a series of classic experiments, demonstrated this speed-accuracy tradeoff for reciprocal (or serial) goal-directed aiming tasks. He found that as the difficulty of the task increases via the decrease of target width (W) and/or increase of movement amplitude (A), so does the movement time (MT) required to complete it, and formally characterized this relationship with the following equation:where “a” and “b” are empirical constants, and log2 (2A/W) comprises the index of difficulty (ID) in bits. This speed-accuracy relationship has proven extremely versatile, leading to its designation as “Fitts’ law” and its widespread application in the field of motor control and human–computer interaction (for reviews, see Mackenzie, 1992, Plamondon and Alimi, 1997).
Although the importance of visual target and effector feedback to motor control has been recognized following the pioneering research of Woodworth (1899), manipulations of these factors have not often been examined with an eye to serial aiming movements and Fitts’ law. Macuga and Frey (2014) compared serial aiming performance with and without effector feedback and found that Fitts’ law was upheld in both cases, but the effect of ID on MT was not as pronounced when effector feedback was removed. However, to our knowledge, no studies have examined how the absence of a target within the context of a serial, goal-directed aiming task influences performance in terms of speed-accuracy tradeoffs. It is also unknown how different forms of confirmatory feedback (i.e., haptic, auditory, etc.) might improve performance under these restricted visual conditions. In the traditional Fitts (1954) aiming paradigm, auditory (metal-on-metal noise) and haptic (felt sensation) feedback may have facilitated faster movement times by decreasing participants’ reliance on visual confirmation for target acquisitions (Glover, 2004, Hatfield et al., 2010). This potentially important confirmatory feedback has been absent in more recent studies on serial aiming, where participants were asked to move a stylus or cursor back and forth on an electronic screen (e.g., Bootsma et al., 2002, Smits-Engelsman et al., 2002). The present study sought to examine these two neglected areas: 1) representation-guided movement and, 2) confirmatory feedback as conveyed through different sensory modalities, at their confluence within a serial aiming paradigm. Online visual feedback is extremely valuable for tasks requiring target acquisition (for a review, see Elliott, Helsen, & Chua, 2001). However, it has been shown that in the absence of visual feedback, movements can be guided reliably based on internal representations (Graziano et al., 1997, Heath and Westwood, 2003), though these internal representations tend to degrade over time (in as little as two seconds for discrete aiming tasks; Elliott, 1992, Heath and Binsted, 2007, Ploner et al., 1998). In serial reaching tasks, visual memory can be used to plan each successive movement but relies more on visual information (if available) once movement execution has begun (Brouwer & Knill, 2007). However, it is unclear how this might play out in an experiment involving rapid aimed movements, particularly with respect to the speed-accuracy tradeoff. In partly addressing this gap, researchers utilized a discrete aiming paradigm to test Fitts’ law under conditions of restricted visual feedback (Wu et al., 2010, Zhang et al., 2015). They showed that Fitts’ law could be applied to visually-restricted movements, though these were more error-prone than visually-guided movements. This may have resulted from the lack of information upon which to base corrective movements. It remains to be seen whether the guidance of serial aiming movements under conditions of restricted visual feedback conforms to Fitts’ law. While discrete and serial movements have produced similar results in tasks with full vision (Fitts & Peterson, 1964) there are reasons why the consequences of target occlusion for a discrete versus a serial aiming task may differ. For discrete aiming, only one target must be held in memory and only for the duration of that single movement. For serial aiming, two targets must be held in memory for a relatively extended period. Thus, serial aiming tasks under visually restricted conditions differ in that there is more than one target that must be internally represented. In the absence of continuous visual feedback, these representations would likely degrade over time, a problem that could be mitigated by confirmatory feedback (Heath & Westwood, 2003).
Though visual feedback (Welch & Warren, 1980) is often dominant, task demands such as target modality (Cameron and López-Moliner, 2015, Sober and Sabes, 2005) or movement direction (van Beers, Wolpert, & Haggard, 2002) can lead to the prioritization of nonvisual feedback. Moreover, recent findings suggest that rapidly accessible, haptic feedback can have a greater influence than visual feedback (Crevecoeur, Munoz, & Scott, 2016). Research on multisensory integration has shown that multimodal cues can improve performance compared to unimodal cues (Angelaki et al., 2009, Ernst and Banks, 2002). Additionally, augmenting visual information with confirmatory auditory (Hatfield et al., 2010) or haptic (Akamatsu et al., 1995, Campbell et al., 1999, Forlines and Balakrishnan, 2008, VercherVitense et al., 2003) feedback can enhance movement-based task performance. Though the role of nonvisual confirmatory feedback has not received much attention, it could potentially facilitate faster movement times with respect to target acquisition.
In the present study, visual feedback of the target was removed, and this continuous feedback was replaced by multiple forms of confirmatory feedback, delivered through individual and paired sensory modalities. To accomplish this, we used a serial movement paradigm resembling the original Fitts’ serial aiming task. The targets of this task were previewed and then removed as each trial began (with no delay between onset of trial and target occlusion), necessitating that participants guide their movements using an internal spatial representation of the targets. Despite this, we hypothesized that each confirmatory feedback condition would conform to the speed-accuracy tradeoff described by Fitts’ law. In other words, the movement times should display the characteristic linear increase in accordance with the index of difficulty. Because participants’ internal representations of the targets may degrade over time, we expected error rates to be somewhat higher than in previous studies where targets were continuously visible. To investigate the influence of confirmatory feedback via different sensory modalities on Fitts’ law, uni- or bi-modal feedback (auditory, haptic, visual, as well as paired combinations of each) was given upon each successful target acquisition. Here, we hypothesized that both visual and nonvisual unimodal confirmatory feedback could be used but that bimodal confirmatory feedback would result in smaller increases in movement times across difficulties (i.e., shallower slopes).
Section snippets
Participants
Participants included 41 undergraduate Oregon State University students. They were recruited using the SONA online system and were given course credit for their research participation (with alternatives for credit available). Written informed consent was obtained. All participants had normal or corrected-to-normal vision and were fluent in English. The experimental protocol was approved by the local Institutional Review Board. A single participant was removed from the results for failing to
Fitts’ law
For all conditions, movement times increased as a function of task difficulty. The linear mixed model regression analysis confirmed that there was a speed-accuracy tradeoff for each sensory feedback condition, suggesting that Fitts’ law applies even under restricted vision (see Fig. 2). The simple regression slopes (b) and index of performance (IP) values for each condition are listed in Table 1. However, the appearance of the data in some of the conditions (most prominently auditory) suggests
Discussion
The present study sought to expand insight into the importance of continuous target presence on the speed-accuracy trade-off described by Fitts’ law. In support of our first hypothesis, Fitts’ law proved robust to visual target removal. The Fitts’ model was able to characterize movements guided by internal spatial representations of the targets, albeit in some instances not perfectly, with some conditions performing substantially worse in the jump to the highest difficulty. A possible reason
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We would like to thank Alexander Piatski and Erik Watterson for their assistance with software development. We would also like to thank Megan Peace, Matt Perella, and Andrea Thody for their assistance with data collection. Finally, we would like to thank Zosia Roberts and Iraiza de Vera for their comments on earlier drafts of the manuscript.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (48)
- et al.
Multisensory integration: Psychophysics, neurophysiology, and computation
Current Opinion in Neurobiology
(2009) - et al.
Informational constraints in human precision aiming
Neuroscience Letters
(2002) - et al.
Target modality affects visually guided online control of reaching
Vision Research
(2015) Intermittent versus continuous control of manual aiming movements
Advances in Psychology
(1992)- et al.
Differential contributions of the superior and inferior parietal cortex to feedback versus feedforward control of tools
Neuroimage
(2014) - et al.
When feeling is more important than seeing in sensorimotor adaptation
Current Biology
(2002) - et al.
Fitts’ law holds for pointing movements under conditions of restricted visual feedback
Human Movement Science
(2010) - et al.
Effect of task-specific execution on accuracy of imagined aiming movements
Neuroscience Letters
(2015) - et al.
Effects of aging on pointing movements under restricted visual feedback conditions
Human Movement Science
(2015) - et al.
Measuring neuromuscular control dynamics during car following with continuous haptic feedback
IEEE Transactions on Systems, Man, & Cybernetics, Part B (Cybernetics)
(2011)
A comparison of tactile, auditory, and visual feedback in a pointing task using a mouse-type device
Ergonomics
Fitting linear mixed effects models using lme4
Journal of Statistical Software
The role of memory in visually guided reaching
Journal of Vision
What you feel must be what you see: Adding tactile feedback to the trackpoint
Multimodal feedback for the acquisition of small targets
Ergonomics
Confidence intervals in within-subject designs: A simpler solution to Loftus and Masson’s method
Tutorials in Quantitative Methods for Psychology
Dynamic multisensory integration: Somatosensory speed trumps visual accuracy during feedback control
Journal of Neuroscience
A century later: Woodworth’s (1899) two-component model of goal-directed aiming
Psychological Bulletin
Humans integrate visual and haptic information in a statistically optimal fashion
Nature
The information capacity of the human motor system in controlling the amplitude of movement
Journal of Experimental Psychology
Information capacity of discrete motor responses
Journal of Experimental Psychology
Separate visual representations in the planning and control of action
Behavioral and Brain Sciences
Multisensory information in the control of complex motor actions
Current Directions in Psychological Science
Cited by (8)
Interaction of hand orientations during familiarization of a goal-directed aiming task
2022, Human Movement ScienceCitation Excerpt :Following familiarization with visual feedback, consolidation of neural connectivity underlies the preservation of error reduction at neural population level (Gallego et al., 2018; Perich et al., 2018). When directional feedback is removed, goal-directed aiming errors magnify and thus rely on feedforward model and proprioceptive feedback to reach a target (Bell & Macuga, 2019; Elliott et al., 2010; Heath & Westwood, 2003). A stored representation of movement planning activates the motor system to maintain the preferred kinematics such as straight-line paths (Shadmehr & Mussa-Ivaldi, 1994), while, proprioceptive input could provide corrective feedback when visual feedback was removed.
Planning lane changes using advance visual and haptic information
2024, Psychological ResearchCorrelates between imagination types and abilities in designing works
2023, International Journal of Technology and Design EducationImproving Effective Throughput Performance using Auditory Feedback in Virtual Reality
2022, Proceedings - SUI 2022: ACM Conference on Spatial User InteractionThe effect of pitch in auditory error feedback for fitts' tasks in virtual reality training systems
2021, Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces, VR 2021