i will take a swing at it... hopefully not just a wild one.  This is somewhat outside my field so i may not use proper terminology..  the concept of state empathy if i am interpreting it correctly is to have empathy for another based on ones own experiences of the emotional state of the other person.  or perhaps more abstractly to be able to assume the mental state of another but only so far as you can relate to their experience.  but even if that is way off i think i may be able to help somewhat. 

Researcher, Lijiang Shen (2010) develops and tests an index of Likert scales for measuring state empathy.  If I am understanding the question correctly you would like to measure the same phenomenon in an observational type study or perhaps even experimentation.  I might begin by trying to reverse engineer Shen's (2010) Survey questions from On a Scale of State Empathy During Message Processing (though there may be better places to start) into a list of measures observable behaviors. 

What this might look like depends largely on contexts.  Are you observing people interacting?  Watching characters on film?  Writing on a blog?  Sitting on a jury?

for example take item 6 from table 1 (pg 508).  (and if you study psychology or social psychology this may be much easier for you that I)  >Item 6 I recognize the characters situation--people often give vocal cues that they have been in similar situations.  or when interacting they may reach out to touch the other person instinctively to indicate their recognition.   or item number 9: when watching the message, I was fully absorbed.  How distracted does the person seem-- in an experimental situation you could bring distraction to bear on interactions to see how easily participants turn their attention to the distraction (here you might want to control for confounding factors--perhaps with some type pre/post test.

or something like that. The real answer to the question is NO i do not know, but it does not seem that the answer, if already out there, is obvious (at least not to me)-- if i'm not way off , I hope this helps

jam

Thanks, Jonathan! I am trying to see if is empathy differs depending on the stimuli you see. For example, if I show you a video of puppies rather than a scary slasher movie, do you feel (in that moment) more (or less) empathic? Or is there no difference? That is not my experiment, but just an example. Thank you for your answer, I really appreciate the time you put into it! 

Hello Karin

There was a similar question asked in the past; I am not sure whether any answers were relevant to you, but it might be worth looking:

https://www.researchgate.net/post/Does_anyone_know_the_most_effective_behavioral_measure_of_empathy2

It might also be worth looking at Prof. Lijiang Shen's paper on ResearchGate:

https://www.researchgate.net/publication/233464674_On_a_Scale_of_State_Empathy_During_Message_Processing

Best wishes

Mary

Article On a Scale of State Empathy During Message Processing

Hi Karin,

Daniel Batson developed a measure of state emotional empathy (or empathic concern) that he used in many of his publications involving an empathy induction. For a few of my studies, I have measured state emotional empathy by utilizing some items from Dr. Batson's measure and combining it with some additional items from the Positive and Negative Affective Schedule (PANAS) to measure sadness, fear, anger, and happiness as a means of determining whether the induction elicited empathic concern (sympathy, compassion) versus a similar emotion experienced by the actor in the stimuli (e.g., sadness). Hope this helps!

Sincerely,

Janelle

Only two suggestions. Not empathy measures per se, but possible behavioral correlates of empathy:  considering as a measure  the agreement of facial emotional displays by source of emphaty and  the subject 

Here are a few behavioral (not self-report) suggestions:

Concordant / Synchronous behaviors (e.g., EDA- http://www.ncbi.nlm.nih.gov/pubmed/23234457; Rhythmic behavior: DOI: 10.1037/a0021302)

Heart Rate Variability / Vagal tone (DOI 10.1007/s10484-011-9161-2; DOI 10.1007/s10578-008-0104-y; http://dx.doi.org/10.1037/pspi0000010)

Adding to the above suggestions:

Measures of brain activity are probably good too. Alessio Avenanti's group has quite a few papers on that topic, and this is a nice review paper (https://www.researchgate.net/publication/224037304_Stimulating_the_brain_to_study_social_interactions_and_empathy) written by Sébastien Hétu, Vincent Taschereau-Dumouchel and Philip L. Jackson.

On another topic, given you want to look at state empathy related to pictures/movies, you may be interested in this review paper by Christian Keysers and Valeria Gazzola (https://www.researchgate.net/publication/260005231_Dissociating_the_ability_and_propensity_for_empathy). My understanding is that we may want to be careful with the way we frame the tasks: just asking people to observe vs. asking explicitly to empathize. Those two ways to frame the task could give us complimentary information by tapping into the person's "ability" to empathize versus the person's "propensity" to empathize. In both cases, you would look at "state" empathy, but seeing if a person's trait empathy influences the response.

While I'm not providing a definite answer to your question, I hope it's helpful and I'm interested in everyone's thoughts about this topic!

Article Stimulating the brain to study social interactions and empathy

Article Dissociating the ability and propensity for empathy

First, you have to distinguish affective empathy and cognitive empathy. "Affective empathy refers to the generation of an appropriate emotional reaction in response to others’ emotions (Feshbach, 1978, 1987). It is comparable to the construct of emotional contagion – the tendency to ‘‘catch’’ emotions from observed emotional states of others. Cognitive empathy is the ability to discern emotional states of others without undergoing emotional contagion."

As Annick Tanguay said, the measures of brain activity would be a good. Some studies use electroencephalographic event-related brain potentials (EEG/ERP) while participants watch emotionally salient stimuli to examine response to the perception of another's distress or pain. 

Per example : Bouchard, S., Bernier, F., Boivin, E., Dumoulin, S., Laforest, M., Guitard, T., Robillard, G., Monthuy-Blanc, J., Renaud, P (2013). Empathy toward virtual humans depicting a known or unknown person expressing pain. Cyberpsychol Behav Soc Netw, 16(1), 61-71

On a Scale of State Empathy During Message Processing

Lijiang Shen Pages 504-524 | Published online: 05 Nov 2010

• https://doi.org/10.1080/10570314.2010.512278