Dear Dr Marchiori,

Perhaps the first experiment on effects of size of portion on intake at a meal was by Jan Fuller in her work with me for a PhD in the mid-1970s.  She went into Clin Psy and had no time to write up for publication and unfortunately both my and the School's copies of her thesis have been borrowed and not returned. 

A complete summary of the findings was published by Viv Lewis and me in Booth, D.A., Fuller, J., & Lewis, V.J. (1981). Human control of body weight: cognitive or physiological?  Some energy related perceptions and misperceptions.  In L.A. Cioffi, W.P.T. James & T.B. Van Itallie (Eds.), The body weight regulatory system: normal and disturbed aspects.  Pp. 305-314.  New York: Raven Press.  That brief report has been cited by some subsequent workers on portion size effects.  The design is more important than the exact numbers and their stats. Some of the published studies captured by your specification do not come up to scratch.

First, participants are never randomised to an experiment.  The recruited sample can only be randomised to conditions - and then only when comparisons between conditions are between subjects. Like most experimenters, we recruited informed volunteers from the staff and students of the university department (who had not previously taken part in an experiment on food).  The design and analysis were within subjects, with sequences balanced across the sample (and no reliable differences between sequences found).

Secondly, the design was a measurement, not just an experimental versus control demonstration, in two senses.  (1) There were two quantative levels of the independent variable (factorial design) - larger and smaller portion sizes.  (2) Any portion size effect was scaled on a variety effect (already known to exist, at least on confectionery), i.e. specificity and sensitivity could have been calculated.

All our experiments include ordinary foods from everyday menus (not buffets) with recorded mouthfuls from a hidden supply ad lib.  This lunch experiment was entirely on quarters of sandwiches with common fillings plus a plain pieces of cake as 'dessert.'  Weighed portions were placed in the trays of an automatic dispenser (Silverstone technique). 

The most important point was partial compensation within the meal - that is, the difference in intake at meals on larger portions and smaller-portion meals was less than the difference cin portion sizes.  In this first experiment we were not interested in the highly complex question of compensation at subsequent meals: there are so many processes that could wipe out a larger lunch later in the day, even if the effect persists on repeated trails (as many effects do not, e.g. of deisguised variations in fat content).

I hope that helps.  - David.

David A. Booth           Messages: +44 121 414 4932
School of Psychology           Fax: +44 121 414 4897
University of Birmingham, Edgbaston, B15 2TT, U.K.
Web:  http://www.birmingham.ac.uk/booth-david
Email: [log in to unmask]
________________________________________
From: The British Feeding and Drinking Group JISCmail list. [[log in to unmask]] On Behalf Of Suzanne Higgs [[log in to unmask]]
Sent: 26 October 2012 17:41
To: [log in to unmask]
Subject: Call for unpublished studies

please respond directly to [log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]>

Dear colleagues,

My name is David Marchiori and I am currently working as a post-doctoral researcher at Tilburg University (Netherlands) under the supervision of Dr. Esther Papies (Utrecht University). We started a project to meta-analyze the studies examining the portion size effect on food intake. We would like to ask whether you have any new studies on this subject (unpublished, in preparation, under review or else).
We are looking for experiments with a randomized assignment of participants, manipulating the amount (weight or volume) of food served (i.e., food portion size is pre-determined by the experimenter) and that measure actual food intake (with a food scale, by counting food items or else).
We would also be very pleased to know if you were aware of any other unpublished studies that were run by any of your colleagues. You can reach me at this email address: [log in to unmask]
Thanks so much in advance!
Kind regards,
David Marchiori

David Marchiori
Post-doctoral research fellow
National Research Fund, Luxemburg

Tilburg School of Social and Behavioral Sciences
Department of Social Psychology
Tilburg University
Warandelaan 2
P.O. Box 90153
5000 LE Tilburg
The etherlands
www.tilburguniversity.edu/research/social-and-behavioral-sciences/