physiological and self-reported data (n=1)? - 22 september 2017 - - - PowerPoint PPT Presentation

▶
physiological and self reported data n 1
SMART_READER_LITE
LIVE PREVIEW

physiological and self-reported data (n=1)? - 22 september 2017 - - - PowerPoint PPT Presentation

Aug 23, 2022 98 likes •631 views

How to analyze a dynamic system of physiological and self-reported data (n=1)? - 22 september 2017 - H.G. van Lier When developing an just in time intervention you try to predict the future for a person. First we need to evaluate if this is

slide-1
SLIDE 1

H.G. van Lier

How to analyze a dynamic system of physiological and self-reported data (n=1)?

  • 22 september 2017 -
slide-2
SLIDE 2

When developing an just in time intervention you try to predict the future for a person.

slide-3
SLIDE 3

First we need to evaluate if this is possible in a context.

slide-4
SLIDE 4

Is there dependence between physiological and self-reported craving ov

  • ver

r ti time?

slide-5
SLIDE 5

Predict self-reported craving with physiological craving?

Is there dependence between physiological and self-reported craving ov

  • ver

r ti time?

slide-6
SLIDE 6

Replace self-reported measurement with physiological measurement? Predict self-reported craving with physiological craving?

Is there dependence between physiological and self-reported craving ov

  • ver

r ti time?

slide-7
SLIDE 7

Replace self-reported measurement with physiological measurement? (Dis)prove dependence between self-reported and physiological craving? Predict self-reported craving with physiological craving?

Is there dependence between physiological and self-reported craving ov

  • ver

r ti time?

slide-8
SLIDE 8

Replace self-reported measurement with physiological measurement? (Dis)prove dependence between self-reported and physiological craving? Predict self-reported craving with physiological craving?

2 physiological 2 self-reported

Is there dependence between physiological and self-reported craving ov

  • ver

r ti time?

slide-9
SLIDE 9

Data

Questionnaire every 3 hours.

slide-10
SLIDE 10

Variables

2 physiological:

  • (mean) skin conductance (SC) level
  • (total) amplitude
slide-11
SLIDE 11

Variables

2 physiological:

  • (mean) SC level
  • (total) amplitude
slide-12
SLIDE 12

Variables

2 physiological:

  • (mean) SC level
  • (total) amplitude

(Leiner, Fahr & Früh, 2012)

slide-13
SLIDE 13

Variables

2 physiological:

  • (mean) SC level
  • (total) amplitude

(Leiner, Fahr & Früh, 2012)

slide-14
SLIDE 14

Variables

2 physiological:

  • (mean) SC level
  • (total) amplitude

(Leiner, Fahr & Früh, 2012)

slide-15
SLIDE 15

Variables

2 self-reported:

  • craving
  • coping
slide-16
SLIDE 16

Variables

2 self-reported:

  • craving
  • coping

How strong is your craving currently? On a scale of 0 (no craving) to 10 (extreme craving).

slide-17
SLIDE 17

Variables

2 self-reported:

  • craving
  • coping

To what extent do you think you are able to resist your craving currently? On a scale of 0 (not resistible) to 10 (easy to resist).

slide-18
SLIDE 18

Cattel’s Data box (Cattel, 1952)

Variables

slide-19
SLIDE 19

N=1

Variables

slide-20
SLIDE 20

Dynamic system

Two or more variables measured over time. Not one outcome and another explanatory variable, but a system

  • f variables continuously influencing each other back and forth
  • ver time.
slide-21
SLIDE 21

Longitudinal data:

  • (Linear) trend
slide-22
SLIDE 22

Longitudinal data:

  • (Linear) trend

Time series data:

  • Autocorrelation
  • (Linear) Trend
slide-23
SLIDE 23

Time series data

We want to study:

slide-24
SLIDE 24

Time series data

We want to study:

  • relationships between a variable and itself on prior time point:

autoregressive relations

slide-25
SLIDE 25

Time series data

We want to study:

  • relationships between a variable and itself on prior time point:

autoregressive relations

  • relationship between different variables on prior time point:

cross-lagged relations

slide-26
SLIDE 26

Physiology Physiology Craving Craving

Vector Auto Regressive Model

TIME

T-1 T

slide-27
SLIDE 27

Physiology Physiology Craving Craving

TIME

Vector Auto Regressive Model

T-1 T

slide-28
SLIDE 28

Physiology Physiology Craving Craving

Vector Auto Regressive Model

autoregressive relation TIME

T-1 T

slide-29
SLIDE 29

Physiology Physiology Craving Craving

Vector Auto Regressive Model

cross-lagged relationships TIME

T-1 T

slide-30
SLIDE 30

Physiology Physiology Craving Craving

Vector Auto Regressive Model

covariance error

TIME

T-1 T

slide-31
SLIDE 31

TIME

Vector Auto Regressive Model

T-1 T

Amplitude Level Craving Coping

slide-32
SLIDE 32

Vector Auto Regressive Model

Two Physiological parameters Two Self-reported parameters

Amplitude Level Craving Coping

TIME T-1 T

slide-33
SLIDE 33

Time series data

Y Y1 Y2 Y3 Y4 … YT

slide-34
SLIDE 34

Time series data

Y Y at lag 1 Y1 Y2 Y1 Y3 Y2 Y4 Y3 … … YT YT-1 YT

slide-35
SLIDE 35

Time series data

Y Y at lag 1 Y1 Y2 Y1 Y3 Y2 Y4 Y3 … … YT YT-1 YT

slide-36
SLIDE 36

Vector Auto Regressive Model

Two Physiological parameters Two Self-reported parameters

Amplitude Level Craving Coping

TIME T-1 T

slide-37
SLIDE 37

Results

Amplitude Level Craving Coping

+ + +

  • TIME

T-1 T

slide-38
SLIDE 38

No dependence between physiology and self-reported craving

  • ver time for this person.

Conclusion

slide-39
SLIDE 39

No dependence between physiology and self-reported craving

  • ver time for this person.

Craving predicts coping 3 hours later and Coping predicts craving 3 hours later

Conclusion

slide-40
SLIDE 40

No dependence between physiology and self-reported craving

  • ver time for this person.

Craving predicts coping 3 hours later and Coping predicts craving 3 hours later Total amplitude predicts mean SC level 3 hours later

Conclusion

slide-41
SLIDE 41

Wrap Up..

If you want to predict the future for a person, it is advisable to use a VAR model (instead of linear regression) to evaluate the dependence between physiological and self-reported measures.

slide-42
SLIDE 42

Wrap Up..

If you want to predict the future for a person, it is advisable to use a VAR model (instead of linear regression) to evaluate the dependence between physiological and self-reported measures. Added benefit: You don’t need to identify an outcome and an explanatory variable, but can analyze a system of variables continuously influencing each other back and forth over time.

slide-43
SLIDE 43

Future research

  • Amount of measurements needed to determine an

individualized just in time intervention strategy?

slide-44
SLIDE 44

Future research

  • Amount of measurements needed to determine an

individualized just in time intervention strategy?

  • Other physiological parameters might predict craving?
slide-45
SLIDE 45

Future research

  • Amount of measurements needed to determine an

individualized just in time intervention strategy?

  • Other physiological parameters might predict craving?
  • Does a similar non-dependence between the physiological and

self-reported parameters exist in other persons as well?

slide-46
SLIDE 46

Future research

  • Amount of measurements needed to determine an

individualized just in time intervention strategy?

  • Other physiological parameters might predict craving?
  • Does a similar non-dependence between the physiological and

self-reported parameters exist in other persons as well?

  • Physiology might predict relapse?
slide-47
SLIDE 47

Questions?

H.G. van Lier h.g.vanlier@utwente.nl

slide-48
SLIDE 48

Significant results only

.445 (.166) .384 (.193) 1.875 (.671) 1.216 (.418) .070 (0.024) .029 (.010) .206 (.078)

Total Amplitude Mean Level Craving Coping

TIME T-1 T

slide-49
SLIDE 49

MPLUS CODE

TITLE: Physiology vs self-reported data; DATA: FILE IS y.dat; VARIABLE: NAMES ARE Crave Crave1 Coping Coping1 Amp Amp1 Level Level1; USEVARIABLE ARE Crave Crave1 Coping Coping1 Amp Amp1 Level Level1; MISSING ARE ALL (999); OUTPUT: TECH1 MODINDICES; MODEL: Crave ON Crave1; Crave ON Coping1; Crave ON Amp1; Coping ON Coping1; Coping ON Crave1; Amp ON Amp1; Amp ON Crave1; Amp ON Level1; Level ON Level1; Level ON Amp1; Amp WITH Crave; Crave WITH Coping; Level WITH Amp ;

slide-50
SLIDE 50

Normal regression

𝑧 = 𝛾1𝑦1 + 𝜗

slide-51
SLIDE 51

Linear trend

𝑧 = 𝛾1𝑦1 + 𝜗 𝑧𝑢 = 𝛾𝑢𝑢 + 𝜗

slide-52
SLIDE 52

Auto correlation

𝑧 = 𝛾1𝑦1 + 𝜗 𝑧𝑢 = 𝛾𝑢𝑢 + 𝜗 𝑧𝑢 = 𝛾𝑢−1𝑧𝑢−1 + 𝜗