Data mining and neurocomputational
modeling in the neurosciences
Kimberly Kirkpatrick
Kansas State University
Behavioral/Cognitive Neuroscience
Mission: Pinpoint the neurobiological mechanisms
that underlie complex cognitive processes and
resulting behaviors
 Growth areas

◦ New techniques, refinements in techniques
◦
◦
◦
◦

 Neuroimaging, optogenetics, cyclic voltammetry
Focus on systems and circuits
Interaction of complex processes
Molecular  Molar (different levels of analysis)
Computational modeling
As a result…
◦ Richer and larger data sets
Links with current funding trends
NIMH: Division of Neuroscience and Basic
Behavioral Science (DNBBS)
 Develop new and use existing physiological and
computational models to understand the biological
functions of genes, gene products, cells, and brain
circuits in normal and abnormal mental function.
 Elucidate how cognitive, affect, stress, and
motivational processes interact and their role(s) in
mental disorders through functional studies spanning
levels of analysis (genomic, molecular, cellular, circuits,
behavior) during development and throughout the
lifespan.
Analytics, Data Mining, and
Neurocomputational modeling

Data mining
◦ The process of knowledge discovery in databases
◦ Data mining should be hypothesis driven
 Surgical approach
 Look for converging patterns

Neurocomputational modeling
◦ Understand computational processes that
underlie complex behaviors
 Deeper insight into mechanisms
◦ Bridge between neurobiology and behavior
 Brain-behavior translation
Three ways of collecting data

Pavlovian conditioning
Tone
Intertrial interval
Food
◦ A tone is followed by food
◦ A response (e.g., rat poking head in food cup) is
measured
Data collection method 1: Record the total
number of responses during the tone.
 Data collection method 2: Record responses
during time bins during the tone.
 Data collection method 3: Record every
response with a time stamp (time-event code)

Responses /
Responses / mi
25
5
Data Collection Method 1:
Record the number of responses during the
tone
20
15
C
10
5
0
0
1
2
3
4
4
3
M
2
S
1
L
0
5
Tone
0
5
Two-session Blocks
Run a counter
35
10-s Tone 12
30
20-s Tone 10
Responses / min .
Responses / min.
14
25
20
15
40-s Tone
10
M
5
S
0
L
1
2
3
Two-session Blocks
10
Time since CS onset (s)
15
Food
Noise CS2
0
Intertrial Interval
4
5
Jennings, Bonardi, & Kirkpatrick (2007)
Noise CS2
Total number
Total time
8
6
4
2
0
0
5
10
15
20
25
Time since CS onset (s)
30
35
Resp
C
M
S
L
4
5
4
3
Data Collection Method 2:
Binned responses
2
1
0
0
5
10
Time since CS onset (s)
15
20
Tone
Noise CS2
Food
10-s Tone
14
Responses / min .
12
M
S
L
4
5
10
Advance Pointer
20-s Tone
8
40-s Tone
6
Run a counter
4
2
0
0
5
10
15
20
25
Time since CS onset (s)
30
35
Jennings, Bonardi, & Kirkpatrick (2007)
40
Data Collection Method 3:
Time-event codes
Time stamp in ms
841.005
1564.005
1650.005
2901.005
3666.005
3856.005
15409.005
19075.005
20331.005
21975.005
47126.006
47277.006
47391.006
47495.006
47598.006
55217.006
55268.006
59765.005
59959.010
60793.005
62070.005
62326.005
62377.005
62411.005
63585.005
64494.005
64882.005
65873.005
66514.005
66741.005
69959.020
69959.013
70059.023
70477.005
106429.005
108570.006
108702.006
109337.010
112883.005
113133.005
119337.020
119337.013
119387.023
120100.005
Event codes
Head entry into food cup = 005
Drinking from water tube = 006
Tone on = 010
Tone off = 020
Food on = 013
Food off = 023
Why time-event codes?
Advantages
Challenges


Flexibility in data
analysis
 May be able to use
one data set for
many purposes
◦ Grant applications
◦ Computational
modeling
◦ Generate new
research questions
Data management
and archiving (5003000 data files/study)
 Requires
programming and
data analysis skills
◦ Data mining

How involve
students?
Techniques and tools

Using MATLAB (“matrix laboratory” developed by The
Mathworks) for data mining
◦ Custom functions and scripts for data extraction
and data reduction
◦ Statistical analysis

Graphical user interface (GUI) using
GUIDE in MATLAB
Techniques and tools
1
Figure window
0.9
0.8
0.7
Parameter Selection
Toolbox
0.6
0.5
0.4
0.3
0.2
0.1
0
0
0.1
0.2
0.3
0.4
Status window
0.5
0.6
0.7
0.8
0.9
1
Approaches to modeling

Traditional approach:
◦ Independent variable  Behavior
relationships
◦ Use metaphors for modeling intervening
processes
◦ E.g., scalar timing theory uses the metaphor of
a stop watch to explain timing behavior

Neural plausibility
Neurocomputational modeling

Develop computational
models that are guided
and constrained by the
properties of the
relevant neural circuitry
Integration/Interaction
Reward
◦ Neural circuitry
◦ Firing dynamics
◦ Neurotransmitter
dynamics

Map onto quantitative
aspects of behavior
Timing
Techniques and tools




Develop model simulations in MatLab for specific
tasks and behaviors
Obtain output and compare with behavior
Vary parameters of the model to improve prediction
of data from individuals
Can create GUIs for model simulations in GUIDE
Summary and Conclusions
 Develop
stronger training programs
for students
 Sharing data
 Sharing tools for analysis and
modeling