• Measuring the speed of nervous system conduction
• And, a tiny lesson in open, transparent, reproducible data science

• How fast does the nervous system conduct information?
• Prior evidence. Last spring's data
  • Proprioception vs. touch
• Why do we care?

We predict that the speed of conduction will be …

• Speed = Distance/Time
• Chain of participants to make distance larger
  • If typical person ~ 1.5 m, then
  • at s=30 m/s, t = d/s -> 1.5/30 = 0.05 secs.

• Squeeze ankle
• ankle_shoulder + shoulder_brain + brain_decide + brain_shoulder + shoulder_hand

• Squeeze shoulder
• shoulder_brain + brain_decide + brain_shoulder + shoulder_hand
• Condition 1 - Condition 2
• ankle_shoulder + shoulder_brain + brain_decide + brain_shoulder + shoulder_hand

• sum(ankle_shoulder) for all participants -> Distance
• mean(time(Condition 1)) - mean(time(Condition 2)) -> Time
• Speed = Distance/Time

• Stop watch
• Tape measure

• Same hand or dominant?
• Alternate ankle/shoulder or one condition before the other?
• How many trials?
  • Fixed number?
  • When reach asymptote?

• Data file with body measurements
  • participant, ankleshoulder (cm)
• Data file with reaction times
  • trial {1…n}, condition {ankle, shoulder}, time (s)

# Load R packages
library("googlesheets")
suppressPackageStartupMessages(library("dplyr"))
suppressPackageStartupMessages(library("ggplot2"))

psych260 <- gs_title("psych-260-fall-2017")

## Sheet successfully identified: "psych-260-fall-2017"

psych260 %>% 
  gs_read(ws = "distance") ->
  distance

## Accessing worksheet titled 'distance'.

## 
Downloading: 100 B     
Downloading: 100 B     
Downloading: 110 B     
Downloading: 110 B     
Downloading: 110 B     
Downloading: 110 B     
Downloading: 110 B     
Downloading: 110 B

## Warning: Missing column names filled in: 'X3' [3]

## Parsed with column specification:
## cols(
##   participant.id = col_integer(),
##   ankleshoulder = col_integer(),
##   X3 = col_character(),
##   comment = col_character()
## )

dist.hist <- ggplot(data = distance, aes(x=ankleshoulder)) +
  geom_histogram(bins = 5)

with(distance, summary(ankleshoulder))

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   124.0   127.0   132.0   132.4   137.0   146.0

# Calculate sum
dist.sum = with(distance, sum(ankleshoulder, na.rm = TRUE))

The total distance is 1192 cm.

psych260 %>% 
  gs_read(ws = "time") ->
  time

## Accessing worksheet titled 'time'.

## 
Downloading: 120 B     
Downloading: 120 B     
Downloading: 120 B     
Downloading: 120 B     
Downloading: 120 B     
Downloading: 120 B     
Downloading: 120 B     
Downloading: 120 B

## Warning: Missing column names filled in: 'X4' [4]

## Parsed with column specification:
## cols(
##   trial = col_integer(),
##   condition = col_character(),
##   time = col_double(),
##   X4 = col_character(),
##   comment = col_character()
## )

# Plot data
time.plot = ggplot(data = time, aes(x=trial, y=time, color=condition)) +
  geom_point() + 
  geom_line()

time %>% 
  filter(condition == "ankle") -> 
  ankle.times

time %>% 
  filter(condition == "shoulder") -> 
  shoulder.times

time.diff <- data_frame(trial=unique(time$trial),
                        time=ankle.times$time - shoulder.times$time)

time.diff.plot = ggplot(data = time.diff, aes(x=trial, y=time)) +
  geom_point() + 
  geom_line()

time.diff$speed <- (dist.sum)*.01/time.diff$time

speed.hist <- ggplot(data = time.diff, aes(x=speed)) +
  geom_histogram(bins = 5) +
  xlab("Speed (m/s)")

speed.plot <- ggplot(data = time.diff, aes(x=trial, y=speed)) +
  geom_point() +
  geom_line() +
  ylab("Speed (m/s)")

speed.plot

• We tested the mean speed of neural propagation in a sample of n=9 college-age adults.
• The mean speed of neural propagation over 5 trials was 36.5637835 m/s with a range of [9.3125, 54.1818182] m/s.
• These findings are generally in accord with values we would expect from the literature.