The hippocampus is involved in memory for recently learned,
but not highly familiar environments
Hirshhorn,
1
1,2
M. ,
Moscovitch,
1,2
M. ,
Grady,
2
C.L. ,
Rosenbaum,
2,3
R.S. ,
Winocur,
2
G .
Department of Psychology, University of Toronto, Toronto, ON, Canada, 2 Rotman Research Institute, Toronto, ON, Canada, 3 Department of Psychology, York University,
Toronto, ON, Canada
Results Continued
Results
Introduction
•There is much evidence to support a role for the hippocampus in learning the
layout of a new environment; however, it’s role in the long-term storage and
retrieval of such memories is disputed.
Comparison Across Sessions
ROI Analysis of the Right Hippocampus
100
90
80
70
0.1
60
•Some studies report hippocampal activation (Spiers & Maguire, 2006) whiles
other studies do not report hippocampal activation during mental navigation tasks
(Rosenbaum et al., 2004)
Session1
50
Session2
40
*
0.08
0.06
30
Session 1
0.04
20
10
Session 2
0.02
0
•The current study aims to test the hypothesis that a spatial representation of a
large-scale environment, that is sufficient to support mental navigation, can exist
independently of the hippocampus. In particular, we wish to test the idea that
spatial representations that are initially dependent on the hippocampus can
become independent of the hippocampus over time.
Dist
Prox
Route
0
Figure 1. Accuracy data for the three mental navigation tasks is shown for the eight
participants who were available for both sessions.
-0.02
1
Figure 4. Mean percentage signal change in the right hippocampus for sessions 1 and 2.
(p< 0.005)
Session 1 (N=13)
Conjunction analysis across tasks
Disjunction Analysis: Activations unique to Session 2
•To test this hypothesis, we compare brain activation during mental navigation
tasks in a large-scale environment, downtown Toronto, at two time-points: once
when participants are newly-arrived to Toronto, and after a year of living and
navigating within the city.
Methods
Participants
Session 1
• Thirteen participants (5 male; mean age 26.7 yrs, SD = 4.0; 2 left-handed) who
had less than one year of experience living and navigating in downtown Toronto
Session 2
•Eight participants from the initial set (1 male; mean age X years; 1 left-handed)
repeated the experiment after one year of living and navigating in downtown
Toronto
Toronto Public Places Test
•Task 1: Distance judgments. Estimate the distance between two landmarks.
(< 2.5 km < ?)
•Task 2: Proximity judgments. Select which of two landmarks is closest to a
reference landmark.
•Task : Blocked-route navigation. Imagine an alternative route between two
landmarks if the most direct route is blocked.
Figure 2. The right hippocampus and left precuneus are activated by all three mental
navigation tasks in session 1.
Region
R middle temporal gyrus
R hippocampus
R fusiform gyrus
R parahippocampal gyrus
L middle temporal gyrus
L superior temporal gyrus
L supramarginal gyrus
R precentral gyrus
L middle frontal gyrus
L inferior frontal gyrus
L superior frontal gyrus
L precuneus
R postcentral gyrus
x
L postcentral gyrus
R cuneus
y
z
53
31
41
-34
-54
-57
-66
45
-25
-57
-21
-11
11
60
-10
15
-53
-23
-36
-14
0
-20
-49
-10
33
27
20
-50
-39
-24
-38
-91
7
-9
-17
-16
-10
-1
25
57
40
4
62
32
70
37
68
32
volume
t-stat
6550
2.77
505
2.77
501
3.45
295
1.75
158
2.20
136
2.21
4623
3.33
1607
2.19
942
2.51
137
2.20
133
2.39
1567
3.00
611
2.19
103
2.33
108
2.23
1698
2.46
Table 1. Regions of common activation in all three mental navigation tasks (p<0.000125)
2 sec
< 2.5 km < ?
Session 2 (N=8)
Conjunction analysis across tasks
+
Museum
Figure 3. The right parahippocampal gyrus and right precuneus are activated by all three
mental navigation tasks in session 2.
Data Analysis
•AFNI, version 2.0 software package
•Head motion correction, physiological motion correction, signal normalization
•Activation maps were transformed to Tailairach coordinates & smoothed with a
Gaussian filter of 6-mm FWHM
Region
R middle frontal gyrus (BA 6)
L middle frontal gyrus (BA 6)
R medial frontal gyrus (BA 10)
R inferior frontal gyrus
L inferior frontal gyrus (BA 47)
R precentral gyrus
R superior frontal gyrus (BA 10)
L superior frontal gyrus (BA 6)
L cingulate gyrus
R middle temporal gyrus
L middle temporal gyrus (BA 21)
R superior temporal gyrus (BA 38)
L superior temporal gyrus (BA 38)
R parahippocampal gyrus
L inferior temporal gyrus
R insula
L insula (BA 13)
L caudate
L superior parietal lobule
R paracentral lobule (BA 6)
x
32
-28
-41
-47
5
9
48
37
-22
41
20
-4
-17
52
-58
48
-47
23
-52
35
-43
-1
-15
-34
8
y
14
20
46
41
66
51
29
35
32
-12
66
27
-33
-32
-27
15
19
-41
-18
14
-5
15
20
-68
-29
z
volume t-value
58
2949
1.52
54
6100
1.47
-4
153
0.7
17
72
0.726
8
2808
0.97
-17
53
1.13
5
255
1.02
-1
79
0.709
-5
11411
1.833
58
865
1.607
14
98
0.826
64
406
1.29
25
1053
0.96
1
541
0.91
-1
697
1.12
-10
56
0.9
-19
247
0.8
0
304
0.72
-21
141
1.1
0
60
0.77
1
314
1.23
7
682
1.54
10
54
0.92
55
746
1.07
66
70
0.71
Table 2. Regions of common activation in all three mental navigation tasks (p<0.005)
TEMPLATE DESIGN © 2008
www.PosterPresentations.com
x
3
-23
y
-66
-58
z
volume t-stat
0
464
0.79
-4
300
1.00
2
59
-4
148
0.74
-21
26
-20
36
36
-98
-5
3
0
136
72
129
1.29
0.54
0.64
Table 3. Brain regions that were active during session 2, but not session 1 (p< 0.001).
Conclusion
•We document a significant decrease in right hippocampal activation during the
performance of mental navigation tasks, as participants gained familiarity with the
downtown Toronto environment over a one year period.
•Consistent activation over time was observed in a network of brain regions
commonly implicated in mental navigation, including the cuneus/precuneus,
middle and superior temporal gyri, and medial frontal cortex.
8 sec
Image Acquisition
•Participants were scanned with a GE Signa 3 Tesla MRI scanner.
•Anatomical: 3D T1-weighted pulse sequence image (124 axial slices, 1.4 mm
thick, FOV = 22 cm).
•Functional: single-shot T2*-weighted pulse sequence with a spiral readout (26
axial slices, 5 mm thick, TR = 2000, TE = 30 ms, flip angle = 30 degrees, FOV =
20 cm).
Region
R lingual gyrus
L parahippocampal gyrus
R medial frontal gyrus (BA
10)
L middle frongal gyrus &
inferior frontal gyrus
R middle frontal gyrus
L cuneus
•This decrease in right hippocampal activation was accompanied by a
corresponding increase in activation in the right lingual gyrus.
1 sec
CN Tower
Figure 5. Activation unique to session 2 is seen in the right lingual gyrus.
•The hippocampus may support detailed representations of spatial information
that are required to guide navigation in an unfamiliar environment.
•As one becomes increasingly familiar with an environment, navigation can be
supported by a coarse representation of major landmarks and their relative
locations, which is dependent on the posterior lingual gyrus.
Contact
Marnie Hirshhorn: [email protected]
References
Rosenbaum, R.S., Ziegler, M., Winocur, G., Grady, C.L., Moscovitch, M. (2004).
Hippocampus 14 (7): 826-835.
Spiers, H.J. & Maguire, E.A. (2006). Neuropsychologia 44 (10): 1674-1682.