Event Title
Paired Serial Rank Order Analysis Provides Clear Evidence of Seizure Effects
Faculty Mentor
John Neill
Major/Area of Research
Physics, Mathematics
Description
A novel data analysis was conducted on data collected in an earlier experiment.
The purpose of the original experiment was to determine the long
term effects of seizures induced in rat pups during the post-natal period on
learning. Subjects: 8 rats had seizures induced with a neurotoxin, flurothyl,
on postnatal days 6-11, and 8 non-seizure control rats. At 90 days of age, rats
were trained for 12 sessions in an go-no go auditory discrimination task.
White noise signaled reinforcement (S+), and a 2 kHz tone signaled extinction
(S-). The rate of responding during 10 second S- trials were compared.
The S- response rates were greater in the seizure rats than the controls.
There was a small but significant difference between the two groups using
a paired t test in the original analysis of 12 sessions. The data were reanalyzed
in the present study by compiling the data into paired serial rank
order (PSRO). Thus, in the first session, a pair of animals, a control animal
and a seizure animal, were selected because both had the lowest number of
S- responses in their group in that session. Then, the next pair of animals
with the second best performance were compared, etc. PSRO analysis was
conducted for every session. A seizure rat almost always performed worse
than a PSRO control rat, even though the rank order of individual animals
often changed. PSRO comparisons show much clearer differences graphically
than group average data that are typically displayed.
Paired Serial Rank Order Analysis Provides Clear Evidence of Seizure Effects
A novel data analysis was conducted on data collected in an earlier experiment.
The purpose of the original experiment was to determine the long
term effects of seizures induced in rat pups during the post-natal period on
learning. Subjects: 8 rats had seizures induced with a neurotoxin, flurothyl,
on postnatal days 6-11, and 8 non-seizure control rats. At 90 days of age, rats
were trained for 12 sessions in an go-no go auditory discrimination task.
White noise signaled reinforcement (S+), and a 2 kHz tone signaled extinction
(S-). The rate of responding during 10 second S- trials were compared.
The S- response rates were greater in the seizure rats than the controls.
There was a small but significant difference between the two groups using
a paired t test in the original analysis of 12 sessions. The data were reanalyzed
in the present study by compiling the data into paired serial rank
order (PSRO). Thus, in the first session, a pair of animals, a control animal
and a seizure animal, were selected because both had the lowest number of
S- responses in their group in that session. Then, the next pair of animals
with the second best performance were compared, etc. PSRO analysis was
conducted for every session. A seizure rat almost always performed worse
than a PSRO control rat, even though the rank order of individual animals
often changed. PSRO comparisons show much clearer differences graphically
than group average data that are typically displayed.