The role of spatial memory and frames of reference in the precision of angular path integration

Abstract

Angular path integration refers to the ability to maintain an estimate of self-location after a rotational displacement by integrating internally-generated (idiothetic) self-motion signals over time. Previous work has found that non-sensory inputs, namely spatial memory, can play a powerful role in angular path integration (Arthur et al., 2007, 2009). Here we investigated the conditions under which spatial memory facilitates angular path integration. We hypothesized that the benefit of spatial memory is particularly likely in spatial updating tasks in which one’s self-location estimate is referenced to external space. To test this idea, we administered passive, nonvisual body rotations (ranging 40° – 140°) about the yaw axis and asked participants to use verbal reports or open-loop manual pointing to indicate the magnitude of the rotation. Prior to some trials, previews of the surrounding environment were given. We found that when participants adopted an egocentric frame of reference, the previously-observed benefit of previews on within-subject response precision was not manifested, regardless of whether remembered spatial frameworks were derived from vision or spatial language. We conclude that the powerful effect of spatial memory is dependent on one’s frame of reference during self-motion updating.

Executing the homebound path is a major source of error in homing by path integration.

“…The results of our simulations, although contrary to some of the assumptions of the encoding-error model, do not completely discredit the model altogether. Evidence suggests that encoding functions may be context-dependent and can be modified by the experience of the navigator (Abdolvahab et al, 2015; Arthur et al, 2012; Chrastil & Warren, 2014a; Klatzky et al, 1999; Petzschner & Glasauer, 2011; Schwartz, 1999; Turvey et al, 2009), in opposition to the assumptions of the encoding-error model. However, in the current study, each participant had their own encoding functions, using a range of values for leg length and turn angle experienced in both the outbound and return paths of triangle completion.…”

“…Angle reproduction tasks have similarly shown underestimations of large angles and overestimation of small angles depending on the range (Becker, Jürgens, & Boss, 2000; Israël, Bronstein, Kanayama, Faldon, & Gresty, 1996; Ivanenko, Grasso, Israël, & Berthoz, 1997; Jürgens et al, 2003; Klatzky, Loomis, & Golledge, 1997; Marlinsky, 1999; Siegler, 2000; Siegler, Viaud-Delmon, Israël, & Berthoz, 2000; Vidal & Bülthoff, 2010). In addition, angle reproduction is influenced by memory and the reference frame used during the task (Arthur, Philbeck, Kleene, & Chichka, 2012), further suggesting that the encoding function is not context free.…”

“…The results of our simulations, although contrary to some of the assumptions of the encoding-error model, do not completely discredit the model altogether. Evidence suggests that encoding functions may be context-dependent and can be modified by the experience of the navigator (Abdolvahab et al, 2015;Arthur et al, 2012;Chrastil & Warren, 2014a;Klatzky et al, 1999;Petzschner…”