Week 7 - Spatial Recognition

Question 1

Who is patient HM?

Answer 1

Patient HM (reported by Scoville and Milner, 1957):

• HM suffered from epilepsy and had medial temporal lobe removed to control the seizures
• Part of the brain removed was HM’s hippocampus
• This produced memory impairments (Antergrade amnesia)
• Amongst HM’s difficulties was finding his way around his new neighbourhood

Question 2

Outline the hippocampus

Answer 2

• First discovered by Lewis in 1923
• It is part of the limbic system
• The hippocampus is made up of the dentate gyrus and different subfields:
  
  • Pyramidal cell layers are important for spatial learning
• Hippocampus is important in spatial memory in lots of species:
  
  • Clarke’s Nutcrackers has the 5^th largest hippocampus relative to body size in birds
  • Black capped chickadees with lesions to the hippocampus retrieved less food caches than controls (Sherry & Vaccarino, 1989)
  • These bird’s hippocampal volume is largest during October when the bird is storing the most food (Smuldders et al, 1995)
• Lateralisation of the hippocampus in humans:
  
  • Damage limited to the right hippocampus causes spatial impairment (Feigenbaum & Morris, 2004)
  • Patient with damage to the right parahippocampal gyrus could only find his room by counting the doors from the end of the hall (Luzzi et al., 2000)

Question 3

How is the Morris Water Maze used to test spatial behaviour?

Answer 3

Morris Water Maze (Morris et al, 1982):

• The basic procedure for the Morris water navigation task is that the rat is placed in a large circular pool (opaque water) and is required to find an invisible or visible platform that allows it to escape the water by using various cues. The rat swims rapidly until they find the platform and escape.
• Rats have to use the cues around the maze to find the platform again.
• There are two types of water maze tasks:

1. Allocentric – this is hippocampal dependent:

• If you release the rat from different cue points around the maze, the rat will learn the position of the platform in terms of the different cues.
• If you then release the rat from a novel position (e.g. moon), it can still locate the platform – the rat has not just learned routes, they have learned the position of the platform based on cues surrounding it.
• This is known as allocentric spatial processing – the location of one object is defined relative to the location of other objects (object-to-object).

1. Egocentric – does not require the hippocampus, requires the entorhinal cortex:

• _​_Teach the rats to find the platform from one position.
• Releasing the rat from a novel position would result in the rat not being able to find the platform – they would use the same movements from their learnt positions.
• Egocentric – represents the location of objects relative to the body.

What are the results to the water maze?

• Allocentric condition – rat finds its way to the platform fairly quickly, they become quicker as they do more trials.
• Egocentric condition – no difference between the rate of the controls and rats that have lesions to their hippocampus.
  
  • Lesions to the entorhinal cortex also affects guidance (McDonald & White, 1995).
• Feigenbaum and Morris (2004) also showed that rats with hippocampal damage were poor

Question 4

What is the evidence of the hippocampus in human experiments?

Answer 4

• Maguire et al. (1998) - showed activity in the right hippocampus in participants navigating around a virtual town
• Maguire, Frackowiak and Frith (1997) - had taxi drivers talk about their routes and monitored hippocampal activity – showed activation in the hippocampus.
• Maguire et al. (2000) – right posterior hippocampus was significantly larger than the controls.
• Maguire et al. (2003) - do people who are good at navigating go into taxi driving?
  
  • No correlation in non-taxi drivers between hippocampal volume and navigational skills
• Maguire et al. (2006) – taxi drivers have bigger hippocampi than bus drivers suggesting is the spatial learning that increases the hippocampi and not just driving.

Question 5

What are place cells?

Answer 5

• Place Cells have been found in the dorsal hippocampus equivalent to posterior in humans (O’Keefe & Dostrovsky, 1971) - place cells are neurons in the hippocampus that fire when the animal occupies a specific location within its environment.
• Place cells take 10 minutes of exploration to become stable (Wilson & McNaughton, 1997)

Question 6

When do place cells fire?

Answer 6

• Based on intended destination - place cells correlate with where animal will be in 120 milliseconds time (Battaglia, 2004)
• Cells would fire based on the rat’s final destination in a double Y maze Ainge (2007) - suggesting place cells were thinking about where it is headed:
  
  • Animals trained to turn left on even trials and right on odd trials. When rats reached the stem of the maze their place cells would start firing suggesting they are planning where to go next.
• Place cells firing reflects where the animal thinks it is rather than where it is (Skaggs & McNaughton, 1998) – a trial involving placing the rat in the south chamber trial after trial. This caused the south place cells to fire. When the rat was placed in the north chamber first, the south place cells fired because the rat thought it was in the south chamber.
• Changing colour of cues changes the rate of place cell firing (Leutgeb, 2005)
• Distal landmarks are more salient than local beacons in setting place cells (Cressant et al. 1997) – the relationship between a cue and a distal landmark is more stable.
• Cells fired when they were in particular place in a maze, but the place was defined by the cues around the maze – the cues were setting off the place cells.
• Foster (1989) - if animal stopped moving, place cells stop firing, but not if it is just motionless – if the rat stopped itself then the place cells would continue to fire, but if we stopped the rat then the place cells would cease to fire. This could be due to the rats predicting where they are going to go next – if we stop the rat, there is no point in them predicting where to go next.

Question 7

What is the interaction between visual and motion cues? (Chen, King, Burgess & O’Keefe, 2012)

Answer 7

• Chen et al. conducted research to determine how external environmental information & internal movement related information effects place cell firing.
• Six young mice were trained to run a linear track & then to navigate their way through the virtual environment.
• Five different visual cues with different textures were present on the side and end walls to help them navigate. These were created using a virtual reality display. Motion cues were created via an air cushion ball. Baseline recordings were taken in the virtual environment before the subsequent experiments:
  
  • The mouse sat on the air cushion ball whilst looking at the screen depicting the virtual environment. The virtual environment enabled the researchers to isolate specific internal & external cues, as well as to provide conflicting cues.
• This study is split into 4 main parts:

1. Training stage - the animal was given time to acclimatise to the research paradigm. Place cell firing is recorded with vision and motion combined. The mice were trained for 3 days to arrive upon a white disk in the virtual environment, and when they did so they received a soy milk reward. The training showed 79% of the cells were identified as place cells at the end of test day.
2. Visual Information stage – to determine the role of visual information in place cell firing, the research paradigm was changed so that the mice could no longer use visual cues to determine location – they were removed from the VE. The mice still had visual stimuli, but it no longer provided information about its location within the VE. If visual information and motion information are used in conjunction with one another, then this should affect place cell firing. Removing the visual cues resulted in an 81% change in the firing patterns of the place cells. This high percentage indicates that visual information from the presence of cues hugely influences place cell firing and, therefore, also navigation.
3. Motion information stage - place cell firing is recorded without motion information (mouse could no longer run on the ball – air cushion was removed) to determine how much this data input affects place cell firing. The mouse sat immobile on the ball while its view changed automatically Removing motion information resulted in 75% of the place cells showing a change in firing pattern. This shows that motion information is also required to cause place cells to fire.
4. Conflict Stage - last part of the experiment looks at what happens when the mice are presented with conflicting visual and self-motion information. The number of steps exerted on the ball in real life, now equated to half the distance the same number of steps previously covered within the VE. This means that if it previously took 20 steps to get to a certain cue, such as to the start of the golden panel, then it would now take 40 steps and so on. The ball-to-virtual movement was halved.

• Results:
  
  • The mice place cells fired at around 70 steps (in between visual and motion cues)
  • Suggests both internal and external cues play important roles in place field stability
  • Internal cues and external cues interact to build up hippocampal spatial representations
  • The hippocampus is not the only neurological structure involved in place field stability

Question 8

Temporal lobe epilepsy affects the hippocampus so, does damage to the hippocampus caused by temporal lobe epilepsy disrupt spatial learning?

Answer 8

Barkas, Redhead, Taylor, Hamilton & Gray (2012):

1. Session 1:

• Training Trials – trained animals and humans to go to a hidden platform
• Probe Trial – platform removed

1. Two-week gap:
   * Probe Trial 2 to see if animal and humans had remembered where the platform was – platform removed

Three different participants in the human condition:

• Sclerosis – damage to hippocampus
• Surgery – already had surgery to try and control seizures
• Control

Results:

• Both surgery and sclerosis patients find platform slower but do eventually find it – affects acquisition of task
• In probe trial 2 - surgery and sclerosis patients perform worse than controls
• Patients with right damage were affected more
• There is evidence that both short-term working memory and long-term reference memory can be impaired in patients with hippocampal damage (Abrahams et al., 1999).
• An inability to retain learned memories over a long period of time, a phenomenon referred to as accelerated forgetting (Blake et al., 2000)

Question 9

So, what is causing the problem?

Answer 9

• Temporal lobe epilepsy - role of new cell formation in learning
• Common held belief that we only lose neurones in adulthood, never gaining new ones – wrong
• There are two areas of neurogenesis:

1. Olfactory bulb
2. Hippocampus (area involved with learning)

Dentate gyrus:

• In dentate gyrus (what the hippocampus is made up of), new neurones develop from stem cells in the sub granular zone (Hilus) to granule cells in the granule cell layer – turns into a neuron when it gets to granule cell layer.

Neurogenesis:

• Properties of new neurons:
  
  • They have lower activation and plasticity threshold leading to preferential recruitment of these cells into functional circuits (Kee et al., 2007) – good at learning new things
  • Over first month over 50 percent of new neurons die
  • Survival positively regulated by enriched environments or behavioural tasks such as spatial learning (Gould et al. 1999)
  • Reduced neurogenesis leads to poor spatial learning (Stubley-Weatherly, Harding & Wright, 1996)

Question 10

Is it the damage caused by the epilepsy or the damage caused by a reduction in neurogenesis?

Answer 10

• Water maze:
  
  • Control group and KA (seziures) have different times in finding the platform – control group do it quicker.
  • An immediate effect on learning and in probe 2 two weeks later.
  • Control group have more cells coming through the dentate gyrus which form new neurons compared to the KA group.
  • Egocentric tasks (don’t need hippocampus for this) – control and KA showed no difference in this task.
• The rats which suffered seizures were poor at learning where the position of the platform was and remembering were the platform was after a few days
• The rats that suffered lesions had less new neurones in Granule Layer

Is reduction of neurogenesis what is causing poor spatial performance?

• The antidepressant fluoxetine is a powerful stimulant of hippocampal neurogenesis (Malberg et al., 2000) and improves memory deficits in patients with mild cognitive impairment (Mowla et al., 2007)

So, would giving rats fluoxetine restore their spatial performance?

• No significant difference found control and rats with seizures who were taking fluoxetine – suggesting neurogenesis is responsible for spatial navigation in the temporal lobe.
• Probe trial 1 – KA are worse then controls and KA with fluoxetine.
• Probe trial 2 – KA with fluoxetine are the same as KA – accelerated forgetting not solved by increase neurons. Accelerated forgetting may be due to damage to the hippocampus.
• Neurogenesis seems to play a role in acquisition of spatial memory, but restoration of neurogenesis does not restore long term memory
• Not clear from results whether problem to do with long term memory is to do with consolidation or recall