N Back Task Min/Max Values


Author
Message
AnthonyFCollinsSussex
AnthonyFCollinsSussex
Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)
Group: Forum Members
Posts: 52, Visits: 250
Hi there!

Hope all is well. I am posting here after much much struggle. I am using the N-back Task for one of my PhD online experiments. I need to know the minimum and maximum possible score of the N-back task script attached (taken from the Inquisit Library). I have also attached a raw excel data file of myself completing the entire task in order to observe how the N-back DV behaves (DV being accuracy rate). I know it is possible to compute the minimum/maximum values of the DV based on key variables such as correct rejections, false alarms etc, I have tried very hard to work this out... but arrive at different values each time! I need to know this information to generate my predictions/statistical model I am using in a pre-reg document. I could seriously do with some help on this.

Many Thanks in advance.

Attachments
NBACK.iqx (255 views, 36.00 KB)
Dave
Dave
Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)Supreme Being (1M reputation)
Group: Administrators
Posts: 13K, Visits: 104K
AnthonyFCollinsSussex - 1/10/2022
Hi there!

Hope all is well. I am posting here after much much struggle. I am using the N-back Task for one of my PhD online experiments. I need to know the minimum and maximum possible score of the N-back task script attached (taken from the Inquisit Library). I have also attached a raw excel data file of myself completing the entire task in order to observe how the N-back DV behaves (DV being accuracy rate). I know it is possible to compute the minimum/maximum values of the DV based on key variables such as correct rejections, false alarms etc, I have tried very hard to work this out... but arrive at different values each time! I need to know this information to generate my predictions/statistical model I am using in a pre-reg document. I could seriously do with some help on this.

Many Thanks in advance.

The computation of the DV is very straightforward, so I'm not sure where the problem is.

values.DV = (values.TotalHits - values.TotalFA)/values.TotalBlocks

where
- values.TotalHits is the number of hits across all experimental blocks,
- values.TotalFA is the number of false alarms across all experimental blocks, and
- values.TotalBlocks is the number of experimental blocks run.

Hits can only occur in target trials. You have 5 target trials per experimental block, so max. 5 hits per experimental block, min. obviously 0.
False alarms can only occur in non-target trials. You have 10 non-target trials per experimental block, so worst case is 10 false alarms per block, best case is 0.

You run 12 experimental blocks, i.e. a total of 60 target trials and a total of 120 non-target trials. Best case is all hits & no false alarms, so

DV = (60 - 0) / 12 = 5

Worst case is no hits & all false alarms, so

DV = (0 - 120) / 12 = -10


Edited 2 Years Ago by Dave
AnthonyFCollinsSussex
AnthonyFCollinsSussex
Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)Partner Member (769 reputation)
Group: Forum Members
Posts: 52, Visits: 250
Dave - 1/10/2022
AnthonyFCollinsSussex - 1/10/2022
Hi there!

Hope all is well. I am posting here after much much struggle. I am using the N-back Task for one of my PhD online experiments. I need to know the minimum and maximum possible score of the N-back task script attached (taken from the Inquisit Library). I have also attached a raw excel data file of myself completing the entire task in order to observe how the N-back DV behaves (DV being accuracy rate). I know it is possible to compute the minimum/maximum values of the DV based on key variables such as correct rejections, false alarms etc, I have tried very hard to work this out... but arrive at different values each time! I need to know this information to generate my predictions/statistical model I am using in a pre-reg document. I could seriously do with some help on this.

Many Thanks in advance.

The computation of the DV is very straightforward, so I'm not sure where the problem is.

values.DV = (values.TotalHits - values.TotalFA)/values.TotalBlocks

where
- values.TotalHits is the number of hits across all experimental blocks,
- values.TotalFA is the number of false alarms across all experimental blocks, and
- values.TotalBlocks is the number of experimental blocks run,

Hits can only occur in target trials. You have 5 target trials per experimental block, so max. 5 hits per experimental block, min. obviously 0.
False alarms can only occur in non-target trials. You have 10 non-target trials per experimental block, so worst case is 10 false alarms per block, best case is 0.

You run 12 experimental blocks, i.e. a total of 60 target trials and a total of 120 non-target trials. Best case is all hits & no false alarms, so

DV = (60 - 0) / 12 = 5

Worst case is no hits & all false alarms, so

DV = (0 - 120) / 12 = -10


Hi Dave

Thank you so much for this, on this occasion my basic math failed me! I was focusing too much on how the DV was changing to the decimal point in the output and it threw my basic calculations somewhat, per usual over-complicating things.

I very much appreciate that.

Yours Sincerely

GO

Merge Selected

Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...




Reading This Topic

Explore
Messages
Mentions
Search