An investigation of image users across disciplines: A model of image needs, retrieval and use



This study investigates the image behaviors of twenty-four participants in four separate user groups (six in each group: archaeologists, architects, art historians and artists). These groups of participants were chosen due to their heavy reliance on images to perform their daily work routines. Two of the image user groups, archaeologists and art historians, are expected to need images for pedagogical and research purposes while the two remaining groups, architects and artists, are believed to need images for inspiration and problem-solving aims. Additionally, the inclusion of these groups of image users will allow for an assessment of image users' behaviors by discipline and underlying needs. This study will identify the critical characteristics of users' image needs, retrieval and use and these characteristics, in turn, will be used to develop a theoretical model to explain users' image behaviors.


Images, visual representations of the world and ideas around us, have become a pervasive presence in the 21st century. Technological advances in the past two decades and the growth of the Internet have accelerated the amount of visual materials available to us and increased our access to images. Although there has been great deal of enthusiasm for the entry of images into the digital realm, research into image users' behaviors has not seen an equal level of support. This is not a surprising situation given that visual materials have traditionally played a secondary role behind that of the printed word (Turner, 1993; Stafford, 1996). For the individuals within four user groups in this study (archaeologists, architects, art historians and artists) images are critical components in the performance of their daily work routines. The needs underlying their image seeking efforts, their image retrieval methods and assessment practices and the ultimate use of the images they find form the core of this study.


The user groups under investigation in this study are understood to have a high need for images in their daily work routines. While several studies have looked into the topic of how critical images are as resources to the disciplines under scrutiny here (Pisciotta et al., 2005; Challener, 1999; Giral, 1998; Sklar, 1995; Busch, 1994; Chidlow, 1991; Gould, 1988), published studies which focus specifically on the suite of image behaviors of the current study's populations are rare. Bradfield (1976) provides the most comprehensive treatment of the subject in his investigation of image collections within institutions of higher education in England. This study examined a variety of aspects such as the subject breadth and image quality needed in collections to support image users, the time users spent finding images and the affective characteristics which influenced users' decisions to use an image collection. Even here, however, image users' behaviors were discussed generally as a means to evaluate the organization and retrieval of information from analog image collections and not as a topic for direct investigation. Another study which looked at image users' behaviors across disciplines is the VISOR I report completed by the Institute for Image Data Research (Conniss, Ashford & Graham, 2000). With the exception of these two works, explications of image users' behaviors are by and large limited in scope with most studies examining a single aspect in the broader spectrum of need, retrieval and use processes associated with images.

The majority of research surrounding image users' behaviors has fallen within the broad topic area of image retrieval. These investigations have examined query formation, the intellectual access to images and issues concerning image relevance. Query formation studies have examined how users shape their queries (Matusiak, 2006; Pisciotta & Copeland, 2003; Batley, 1988; Bradfield, 1976), the mediation and modification of queries (Jörgensen & Jörgensen, 2005; Goodrum & Spink, 2001; Armitage & Enser, 1997; Enser, 1993; Bradfield, 1976), keyword vs. browse searches (Matusiak, 2006; Cunningham & Masoodian, 2006; Goodrum, Bejune, & Siochi, 2003; Frost et al., 2000; Hastings, 1999; Frost & Noakes, 1998;

Angeles, 1998; Batley, 1988; Bradfield, 1976) and the influence of subject expertise and/or experience on the query process (Matusiak, 2006; Frost et al., 2000; Frost & Noakes, 1998). Much of the research into the intellectual access to images has focused on identifying and examining the useful conceptual access points for images (Jörgensen & Jörgensen, 2005; Cunningham, Bainbridge, & Masoodian, 2004; Jörgensen, 2003; Frost et al., 2000; Jörgensen, 1998; Armitage & Enser, 1993; Roberts, 1985; Bradfield, 1976) and how these may vary according to context or user group (Chen, 2007; Armitage & Enser, 1997; Ornager, 1995). Other work has concerned the challenges surrounding providing intellectual access to images (Chen, 2007; Beaudoin, 2007; Attig, Copeland & Pelikan, 2004; Grout et al., 2000; Bradfield, 1976). Relevance, how users assess and select images to meet their needs, has been examined in several studies. These discuss the criteria used in the evaluation process and/or the impact of users' needs (Choi & Rasmussen, 2003; Goodrum, Bejune, & Siochi, 2003; Choi & Rasmussen, 2002; Efthimiadis & Fidel, 2000; Bakewell et al., 1988; Bradfield, 1976).

While each research study provides an additional piece of information that expands our knowledge surrounding this topic, none presents an overall view of users' needs, methods of image retrieval, and use of images. An assessment of the limited state of research surrounding searches for and use of images is alluded to by Fidel (1997) who notes plainly that several fundamental issues in image retrieval have been ignored.1 It appears that the lack of high-level image user studies which might provide answers to these fundamental questions is due to the absence of theoretical foundations upon which to base research. Chen (2001) states that “[i]nadequate research in user studies and a lack of a theoretical background for the design and evaluation of image databases have been repeatedly mentioned as problems in the image retrieval field.” This study will attempt to develop a theoretical framework with which to view the entire range of image behaviors.

Research Questions

Q1. What information needs drive users to seek images?

Q2. How do users retrieve images?

Q3. How are images used to support users' work tasks?

Research Methods

This study will examine twenty-four participants in four separate user groups (six in each group). These participant groups were chosen because of their heavy reliance on images of cultural objects in the completion of their work tasks. They consist of:

  • Archaeologists - teacher / researchers working within an academic setting.
  • Architects - designers of architectural structures and spaces.
  • Art historians - teacher / researchers working within an academic setting.
  • Artists - creators of works of art.

The choice of these groups of participants creates a situation where two groups (archaeologists and art historians) are expected to need images for pedagogical and research purposes while the two remaining groups (architects and artists) are believed to need images for inspiration and problem-solving aims. These different disciplines and underlying needs will allow for an examination into the possible causes of variation in image users' behaviors. The participants will be selected using a convenience sample.

Several data collection techniques will be employed in this primarily qualitative research study. The first of these techniques is a survey which will be used to gather preliminary data about the participants' demographics, their image needs, their methods of image retrieval, and their use of the visual materials they find. The data collected here will be analyzed using descriptive statistics and this information will begin to form a picture of image users' behaviors. Furthermore, the survey will be useful to the development of the interview questionnaire by providing preliminary data on image behaviors. Semi-structured interviews will be conducted with each participant using audio-tape and transcription techniques. The interviews will allow direct inquiries of the participants so that useful data can be gathered to answer the specific research questions under investigation. Through the interviews the individual experiences and intellectual processes underlying the image needs, retrieval methods and image use can be explored. Themes which emerge from the interviews will be used to discuss and clarify the phenomena surrounding image users' behaviors. The use of the survey and interview methods together will help ensure that rich, descriptive data are collected for the study.

Study Objectives

  • 1. Provide an overview of the current state of research on image needs, retrieval and use.
  • 2. Identify and interpret critical characteristics of users' image needs, retrieval and use.
  • 3. Categorize similarities and differences in image seeking behaviors among user groups.
  • 4. Develop a theoretical model for image seeking behaviors.

Expected Results

The user groups are expected to differ on several dimensions. These dimensions include: the resources and formats used for their image needs; the search strategies they employ; the rigor of the criteria used in relevance; the time and effort expended in the various processes associated with images; the perceptions of their success and frustration in image retrieval and use; and the technological skills they possess. Beyond this it is anticipated that image users search multiple locations for their images with the practice-based user groups generally preferring printed images for their work tasks. A final expected finding is that image searching and image preparation for teaching and research requires an enormous amount of time and support.

  1. 1

    The overlooked and fundamental questions mentioned by Fidel are: What are the differences between image and text retrieval? What image attributes are important for retrieval? What are the characteristics of users' queries for images?