Discovery Lab Answers

Marilyn J. Schlitz, PhD and the joint remote-staring experiments with Richard Wiseman

Marilyn J. Schlitz, PhD is a parapsychology researcher whose remote-staring work is best known for a series of adversarial-collaboration studies run jointly with the skeptic Richard Wiseman. The premise was straightforward: Wiseman’s own remote-staring studies had produced chance results while Schlitz’s had produced statistical significance, so the two ran experiments together — same protocol, same equipment, same participant pool, at the same site — to find out why.

Coverage: this reflects the studies currently in the ESP-Nexus library, not the complete published literature on this question. We have not measured how much of that literature the library holds, so treat what follows as a summary of what we hold rather than a settled account of the field.

Experiments

The retrieved sources and the site’s synthesis identify three joint studies:

• Wiseman & Schlitz (1997) — the first joint study, using the electrodermal-activity (EDA) remote-staring paradigm, with each researcher running a separate arm of trials.
• Wiseman & Schlitz (1999) — an attempted replication of the 1997 design, run with the same basic protocol; this is the paper retrieved as source, which describes replicating “the findings of our initial joint study.”
• Schlitz, Wiseman, Watt & Radin (2006) — a third collaboration extending the paradigm, which the site’s synthesis describes as failing to replicate the experimenter-pattern of the first two.

The distinguishing design feature across all three was the paired-experimenter structure: Schlitz (the proponent) and Wiseman (the skeptic) ran their own participants under the same conditions, so that any divergence would attach to the experimenter rather than to the protocol.

Methodology

The paradigm measured electrodermal activity while a participant was covertly observed (staring epochs) versus not observed (nonstare epochs). In the 1999 study, a “stare”/”control” cue was displayed to the starer, and the program inserted a randomized gap of one to five seconds between periods to disrupt any physiological periodicity in the participant. Trials for the two experimenters were run at the same time and in the same location with identical equipment. The 2006 study used the same basic EDA-staring paradigm and, per the site’s experimenter-effects page, was extended toward examining whether the divergence would generalize to other experimenters differing in belief orientation.

A methodological point Schlitz herself has emphasized (per the site synthesis) is treating the experimenter as a variable to be studied rather than noise to be averaged away — and she notes that the field’s small size and funding constraints make it hard to accumulate the large, multi-lab datasets needed to characterize experimenter effects as a moderating variable.

Data

The audited evidence rows below carry only the paired-arm figures from the three studies; they are reported in different, non-comparable metrics (some as standardized effect sizes with z, some as z-scores alone), so I describe the pattern rather than pooling anything into a single figure.

The pattern the rows show:

• 1997 and 1999: Schlitz’s arm reached significance while Wiseman’s arm did not. Note the direction change the site synthesis and source both flag — in 1999, Schlitz’s participants showed EDA that was higher during nonstare than stare periods, i.e. the significant effect ran opposite in direction to the 1997 result.
• 2006: both retrieved conditions were null, consistent with the site’s description of the third study as a non-replication of the experimenter pattern.

Honest unsettled signals in this evidence set: the reported directions disagree across studies (some positive, some null), and the majority of the rows here — five of the paired arms — are null.

Skeptical critiques

More broadly, Wiseman’s own arms of the joint studies produced consistent null results under an identical protocol, which is itself the strongest skeptical datum: the same procedure that gave Schlitz significance gave the skeptic nothing.

What the experimental data show. The 1999 paper reports that the divergence was real within each study — Schlitz’s participants significant, Wiseman’s not — but the site synthesis and the study record are explicit that **the difference between the two experimenters was not itself statistically significant in either the 1997 or 1999 study.** That matters: the effect that reaches significance is the within-arm stare/nonstare contrast, not the experimenter contrast, so the “experimenter effect” is an inference from a pattern of two significant and two null arms rather than a directly significant experimenter × condition interaction.

Analysis. The three studies present a specific, documented sequence: two collaborations (1997, 1999) in which Schlitz’s arm reached significance and Wiseman’s did not, with the significant effect reversing direction between them, followed by a third (Schlitz, Wiseman, Watt & Radin 2006) in which the experimenter pattern was not reproduced. Schlitz, Wiseman, Watt and Radin (2006) themselves frame two competing readings in the paper: (1) the first two studies captured a genuine effect disrupted in the third by some unidentified feature of the new design or context; or (2) the first two were chance findings or subtle artifacts and the third’s null is the more accurate picture. Independent replication of the experimenter divergence outside this collaboration is not established in the retrieved sources. The between-experimenter difference’s failure to reach significance in the first two studies means a straightforward replication logic is difficult here — a null from a new lab does not by itself adjudicate between those two interpretations.

For the full worked treatment, see the experimenter effects spoke and the remote-staring detection spoke.