A neural wayfinding mechanism adjusts for ambiguous landmark

information

Report: Zhang Jianfeng

Background

• A neural mechanism supporting successful wayfinding needs to distinguish between helpful spatial information and ambiguous or misleading information

• Hippocampal and parahippocampal region are involved in spatial memory and navigation

• parahippocampal gyrus not only responds to scenes but also to the recognition of single objects

Aim

• Find increased activity in the parahippocampalgyrus for objects placed at a decision point only once as compared to objects placed at non-decision points

• Find a neural wayfinding mechanism which adjusts for ambiguous information

Materials and methods

• Participants

– Twenty healthy human adults

– right-handed

– Mean age was 24.75 years

Materials and methods

• Paradigm: two parts + a practice

– A practice task trained in a maze not used in the experimental study task

– a study phase outside the scanner

– a recognition phase in FMRI

Materials and methods

Materials and methods

• Part 1 instruction:– You have been hired as a staff member at the museum for

photography. Your boss wants you to get acquainted with the photo-archive, and to report what material is suitable for the next exposition.

– This will be a photo-exposition on the favourite objects of famous people. You are now sent to the archive that that consists of four floors and holds two copies of each photo.

– Your assignment is to walk through the archive, to learn the route and to get acquainted with the photos. Afterwards you should be especially familiar with photos that are interesting for children (like toys). Find the way by following the photos. You can walk by pressing the four arrow buttons on the keyboard as learned in the training phase.

Materials and methods

• Conditions of part 1(3*2)– each object appeared twice at two different

decision points (D-D objects)

– at two different non-decision points (ND-ND objects)

– once at a decision and once at a non-decision point (D-ND objects and ND-D objects)

– Toys(attention)

– non-toys(no-attention)

Materials and methods

• Part 2: a recognition phase in FMRI

– performed a simple object recognition task

– press either a yes or a no response key

Results

• Behavioural results(accuracy data)

– a non-decision point and a decision point (10.6%; t19 = 3.34, P<0.01)<ND-ND decision point errors (15.1%) <D-D decision point (19.0%; t 19 = -2.72, P<0.01)

– toys (15.9%) < non-toys (22.0%)

Results

• Behavioural results(Response times)

– ND-ND decision(mean 859ms) < D-D(871ms)<N-ND/D-ND(mean 884 )

– Toys(858)< non-toys(891)

Results

• FMRI result:

– To investigate effects of navigational relevance of object location: (right posterior parahippocampalgyrus)

• T-contrasts showed significantly higher beta weights for ND-D and D-ND compared to ND-ND objects

• ND-D and D-ND > D-D

Results

• right middle frontal gyrus:(an area involved in executive functions like cognitive control )– D-D objects > ND-ND objects

• D-D objects > ND-ND objects:– no activity in the medial temporal lobe including the

parahippocampal gyrus was observed

• different turn > same turn – (higher beta weights in the right middle frontal gyrus )

Results

Discussion

• parahippocampal gyrus:

– D-ND vs ND-ND: higher

– D-D vs ND-ND: no difference

– D-ND vs D-D: higher

– independent from paying attention to a specific object category

– reflects automatic processing of navigationally relevant objects

Discussion

• right middle frontal gyrus

– D-D vs ND-ND: Higher

– involved in spatial working memory, plays a special role in monitoring

– Different Turns vs Same Turns: the middle frontal gyrus active stronger

Discussion

• right fusiform gyrus

– Toys vs non-toys: increased neural activity in ventral occipito-temporal regions

– related to paying attention to objects

– not affected by navigational relevance