To my mind, Gramann's research showed at least two important things. Firstly, it found that a difference in how people think about their location in space leads to differences in neural activity exactly when the person is engaged in the spatial computations associated with orienting him- or herself in that space. Secondly, he showed additional evidence for the utility of source localization and ICA-based analyses of human behavior.
One of the chief differences between the "turners" and the "non-turners" was a smaller decrease in 10 Hz Alpha activity (I'm not sure which group had such a decrease). If alpha is an "idling" rhythm, does this mean that one group found it "easier" to do the computation than the other (that it required less cognitive effort/attention)? Since the difference wasn't apparently related to spatial ability, the answer would at first appear to be "no" ... but perhaps this might be investigated using an adaptive paradigm that tried to equalize alpha across both groups (perhaps by providing a more difficult task to the group that showed less of a decrease in alpha?)

I am in complete agreement with Klaus that traditional neuroimaging experiments are hampered by requiring participants to lie or sit still in a scanner, rather than moving about in the world. Related to this, he emphasized that the tunnel paradigm only provides people with visual information about space, whereas real movement also involves, for example, vestibular and proprioceptive information. One question that arises from this is: if we use a variety of information sources to update our spatial representations, precisely which kinds can be produced during imagined as opposed to real movement? For example, when viewing the scenes in the tunnel paradigm, it is possible that efference copies of motor commands for making those movements are produced, which can be an additional input for the updating of spatial representations beyond the visual information provided. Can other sorts of modality specific information, such as vestibular information, be imagined, i.e., produced through top-down processes, to aid in the construction and updating of spatial representations? Or do some information sources only work in a bottom-up fashion?

Given that the paradigm produces no actual rotations or translations, the "ground truth" interpretation would seem to require that the subject be facing forward (no rotation) while the tunnel moves around them (no translation). This would imply that there are four possible interpretations rather than two. These might be labeled "movers / turners" [A], "movers / non-turners" [B], "non-movers / turners" [B], and "non-movers / non-turners" [A] with the answer of A or B thus being relative to both the perceived translation and rotation. Of course, the verbal instructions might override the two "non-mover" interpretations, as might, the bias to interpret wide field optic flow as self-motion. I don't suppose anyone else made a "non-mover" interpretation?

One of the things that I found most fascinating about this talk was that Gramann had no good accounting for the bias to be a "turner" or "non-turner". The absence of any obvious correlates seems to be somewhat startling. If we assume that there is some basis for why some people are pre-disposed to understand the tunnel from an ego-centric perspective and others from an allo-centric perspective, investigating that basis might lead to some interesting insight about how these two systems work together. Along those lines, I would be interested to see whether the relatively consistent 50/50 split that is seen with the virtual tunnel would remain with physical movement. My assumption would be that the physical movement would bias subjects toward ego-centricity, but it would certainly be interesting to compare the characteristics of the subjects in the "turner/non-turner" categories across the virtual and real-world conditions.

I thought Gramann's lecture gave a great overview of spatial cognition, and raised a number of questions related to the topic. Just like Adam Fouse, I'd like to hear more about the turner/non-turner distinction. In addition, I think it would be very interesting to look at the differences between men and women, which Gramann touched upon briefly.

For someone with little background in spatial cognition (like me), his presentation was a good overview of the subject and led to a few questions. Although it may not be directly related to this line of research, I would like to know about the gender differences that he referred to as for subjects and explore more the underlying reasons for these differences.

The distinction between egocentric and allocentric spatial representations is interesting because it raises the question as to whether the neural circuits underlying spatial cognition are also distinct and possibly separate in this manner. Although it has been shown that a preference for one particular method or strategy (egocentric vs. allocentric) often occurs, we discussed that it definitely may still be the case that humans use both simultaneously. There is great evidence to suggest that the retrosplenial cortex is in an ideal location to be involved in spatial cognition, but the exact circuitry that may suggest its participation in two distinct (and arguably separate) forms of spatial cognition would be really interesting to explore further. Specifically, which circuits could define two distinct forms of spatial representation (or navigation) and how anatomically might the retrosplenial cortex act as a integrator of the two (as suggested by the lecturer). In summary, I would be interested in hearing more evidence that egocentric and allocentric spatial representations should be considered distinct in the brain.

One day last week I had to figure out how to get to the Salk Institute from where I was on campus. All this talk about spatial cognition made me particularly aware of my mental mapping of campus. Was I using an allocentric or egocentric reference frame, or a hybrid of the two? How does conscious awareness and experience make use of and modify our inherent tendencies to use one strategy or the other? And how does that change depending on the temporal or spatial context? During Klaus' talk, I couldn't help but wonder if there is any connection between allocentric and egocentric spatial cognition and allocentric and egocentric world views.

I was most interested in a brief slide shown toward the beginning of the talk that described how models of spatial learning divide the development of spatial knowledge into three stages (landmark, route and survey knowledge). While the distinction between allocentric and egocentric spatial representations is one that I would never have considered if not for Klaus' talk, I felt that his explanation of how we build up mental maps perfectly describes the conscious experience of figuring out how to get around in a new location. I can distinctly remember how, the first two or three times I visited places that I later got to know very well - my high school, my college, UCSD - the campus layout seemed somehow qualitatively different than it did later, in a way that's difficult to explain: things just seemed to fit together in a different way. I wonder now if that sensation reflected the fact that I was relying on a form of spatial knowledge that I quickly outgrew.