Introduction

Weather is an important factor for the use and the perception of outdoor urban places (Zacharias et al. 2001; Nikolopoulou et al. 2001; Thorsson et al. 2004; Eliasson et al. 2007). It is, for example, well known that designing outdoor urban places with climate in mind may result in health, social, economical, and environmental benefits (Givoni 1998; Eliasson 2000; Johansson and Emmanuel 2006). In the last two decades, several researchers have also called attention to the psychological variables involved in outdoor place and weather assessment (e.g., Höppe and Seidl 1991; Höppe 2002; Nikolopoulou et al. 2001; Nikolopoulou and Steemers 2003; Nikolopoulou and Lykoudis 2006; Thorsson 2003; Thorsson et al. 2004). In line with this, some findings have recently indicated that the processes of weather assessment may be intertwined with psychological and cultural processes (Knez and Thorsson 2006, 2008) rather than fixed by general thermal indices as suggested by the physiological heat balance models (Höppe 1997).

Concerning the outdoor thermal perception (thermal comfort), more than 100 different indices have been developed over the years signifying the heat exchange between a human body and its surrounding environment (Jendritzky et al. 2001). These are classified into empirical indices [e.g., discomfort index (Thom 1959), apparent temperature (Steadman 1979), wind-chill index (Steadman 1971)] and rational indices based on the human energy balance [e.g. Predicted Mean Vote (Fanger 1970), Physiologically Equivalent Temperature (Mayer and Höppe 1987), Outdoor Standard Effective Temperature (Pickup and de Dear 1999)]. However, none of these over 100 indices, nor the new Universal Thermal Climate Index (UTCI) (www.utci.org ), which is at the moment in progress by an initiative of the International Society of Biometeorology, take into account the psychological variables involved in outdoor thermal perception.

Given all this, the general aim of the present paper was to clarify in some more detail the psychological mechanisms involved in outdoor place and weather assessment, by proposing a conceptual model suggesting direct and indirect links of influence in an outdoor place–human relationship. This work is a result of an interdisciplinary research collaboration comprising scientists from the fields of climatology, psychology, and architecture with an ambition to integrate and enlarge the knowledge of the complex climate–human behaviour link and its implications for sustainable urban design (Knez 2005, 2006; Knez and Thorsson 2006, 2008; Thorsson et al. 2007; Eliasson et al. 2007).

Psychology and environmental assessment

Nikolopoulou and co-authors have, in a series of articles (Nikolopoulou et al. 2001; Nikolopoulou and Steemers 2003; Nikolopoulou and Lykoudis 2006), suggested that psychological adaptation may account for some of the unexplained variance between objective and subjective thermal comfort assessments, comprising parameters such as: naturalness, expectations, experience, time of exposure, perceived control, and environmental stimulation. One of these variables, “experience”, is of particular significance for the understanding of human parameter, involved in not only thermal comfort estimations and psychological adaptation, but in environmental assessment in general; because this concept relates to the central part of a psychological explanation. In their attempt to clarify, in some more detail, this significant psychological variable, Nikolopoulou (et al. 2001) rightly connected “experience” to the cognitive concepts of “memory” and “schemata”, but they wrongly attributed these two concepts to the short-term and long-term experience, respectively (e.g., Sternberg 2006).

Contemporary cognitive psychology (e.g., Eysenck and Keane 2005; Sternberg 2006; Smith and Kosslyn 2007) tells us that when we learn, remember, think, imagine, and reason, a chain of information processing from the detection of a physical stimulus (seeing, hearing, smelling, tasting, and touching) to the conscious experience of it is involved. This sequence of events associates several structures; such as the sensory system (a sensory store where the physical energy reaching our senses converts to the chemical energy and is briefly held; 250 ms to 4 s), CNS (central nervous system; where information is further processed and coded) and mind/brain (part of the CNS comprising memory systems where information is transformed into, for us, cognitive and meaningful information).

The human memory comprises two basic stores, a short-term memory and a long-term memory. The former is of a very limited capacity where the sensory data is cognitively identified and named and held for about 12 s, longer if rehearsed and elaborated (linked to the long-term memory stored knowledge and experiences). The latter store is virtually unlimited, holding information over extremely long periods of time. We could say, in other words, that the short-term memory is the psychological “now” where any distraction of coding, storing, and recalling the ongoing, incoming sensory information (and/or information from the long-term memory) causes the forgetting. The long-term memory, on the other hand, is the psychological “past” that gives us the meaning of the world, because this is the store where all our experiences and knowledge (from facts to personal related events) are stored. The long-term memory is organized into several different memory systems storing declarative (factual knowledge and experiences of which we are directly aware) and procedural (know-how skills such as how to walk and bicycle) types of information.

Furthermore, the declarative parts of the long-term memory contain different organizing structures such as schemata. These are packets of information, mental representations, about different events such as how to behave when playing football, attending a conference, or eating at a restaurant. Schemata also comprise “expectations” related to these events; for example, how we should be served by the restaurant staff. Consequently, this means that a schema may affect our processes of perception and comprehension (e.g., Bartlett 1932; Piaget and Inhelder 1973; Minsky 1975) because it serves both as a guide for action and as a framework for interpreting the world. Finally, this implies that a schema can be viewed as a type of attitude (Eagly and Chaiken 1993), a reasoning that has also been extended to the study of environmental assessment.

Brewer and Treyens (1981) suggested that schemata might affect encoding, storing, and retrieving information in memory for places. In line with this, Knez and Thorsson (2006, 2008) showed that both collectivistic attitudes (Japanese vs Swedish culture that imposes different environmental attitudes upon their members) and individualistic ones (self-related environmental attitude: open-air person “find pleasure in the sea, the woods, the nature” vs urban person “find pleasure in the street-life, the shops, the amusements of the city”) affected perceptual and emotional estimations of outdoor urban places. For example, and concerning the thermal perception (thermal comfort), they suggested, in line with Triandis (1994), that Swedes and Japanese might have evolved different culture-related subjective scales accounting for the different assessments of thermal comfort between the members of the two cultures. Also, Knez (2005, 2006) reported differences in place-related identity as related to the environmental attitude of open-air versus. urban people and some influences of Nordic culture on autobiographical memory (personal experiences and events) for outdoor places.

In sum, and as pointed out by Nikolopoulou and co-authors, the processes of human memory, encompassing “expectations” and “experiences” of outdoor settings, may account for some of the unexplained variance between objective and subjective environmental assessments. Yet Nikolopoulou and Steemers (2003) wrongly divided “experience” into short-term experience related to “memory” and long-term experience related to “schemata”; also, Nikolopoulou et al. (2001) erroneously attributed some of their results to the short-term memory (e.g., Eysenck and Keane 2005; Sternberg 2006; Smith and Kosslyn 2007) when they discussed effects that lasted for 6 days. Finally, it must be noted that “experience” and “expectations” as part of the long-term memory are not only linked to the psychological adaptation, as suggested by Nikolopoulou and Steemers (2003), but also to cognitive features that all kinds of “top-down” psychological processes (driven by high-level cognitive goings-on, existing knowledge, and prior expectations) are based upon and related to (Eysenck and Keane 2005).

Objectives

Given the reasoning outlined above, the conceptual objective of the present article was to suggest a conceptual model, tentatively proposing direct and indirect links of influence in an outdoor place–human relationship (see "Conceptual Model" below). In addition, and in line with this broader analysis of human responses involved in outdoor place and weather assessment, the empirical objective was to investigate the impact of weather and personal factors on participants’ perceptual and emotional estimations of outdoor urban places (see "Empirical Study" below) and by that means test some of the assumptions proposed by the conceptual model.

Materials and methods

Conceptual model

As can be seen in Fig. 1, we suggest a tentative model of direct and indirect influence of place-related parameters on human responses. This conceptualization involves three main organizing entities, namely: place, moderator/mediator and human response.

Fig. 1
figure 1

Proposed direct and indirect (via moderator/mediator) influences of a place on a person (human response)

Full size image

We use place instead of space (Eliasson et al. 2007) because the concept of space (implying only physical and spatial connotations) has over the last decades been redefined by the notion of place, thus including also the psychological and social aspects of space experience (Canter 1997; Graumann 2002). In the words of Canter (1997, p. 118): “…the experience of places evolves out of transactions between the personal, cognitive-emotional system that a person brings to any setting and the socially structured patterns of actions that occur within that setting as given significance by the culture in which the transactions occur.”

Place

Place entails physical base, weather and function. Physical base comprises form (structure, openness), material (surface characteristics), naturalness (degree of artificiality), and ___location (space dimension). Weather includes meteorological parameters such as temperature (air and surface temperature), radiation (direct and reflected long-wave and short-wave radiation), wind (speed, direction, turbulence), humidity, and cloudlessness (cloudy-clear sky magnitude). Function is divided into physical activity (standing, sitting, lying, walking, running, etc.) and social activity (talking, playing, associating, etc.).

Moderator/mediator

Moderators are variables, such as personal factors, that might interact with place parameters in an effect on human responses in a place–human relationship (Evans and Lepore 1997). Consequently, a moderator can be treated as an independent variable to test for an interaction effect between a place variable and a personal factor (moderator) on a human response. Mediators, on the other hand, are variables that intervene between independent and dependent ones; meaning that an independent variable influences a mediator which, in turn, influences a dependent variable (see Evans and Lepore 1997 and Baron and Kenny 1986 for details in how to test mediating effects).

As can be seen in Fig. 1, we have listed three types of potential moderators/mediators: Culture (rules, norms, values); Person, which includes psychological parameters (knowledge/experience, attitude/expectation, belief/preference, perceived control), demographic variables (age, gender, education, etc), biological base (e.g., metabolic rate—energy output needed for bodily functioning—that might be important for thermal sensation and perception/comfort), physical and social activity (these parameters define the function of the Place as well as the Person-related activities); Situation (length of exposure, momentary clothing).

Culture is: “The system of information that codes the manner in which people in an organized group, society or nation interact with their social and physical environment.” (Reber 1985). A member of a culture learns its rules, norms, and values which s/he then shares with other members of the “system”. In addition, culture provides: “…standards for perceiving, believing, evaluating, communicating, and acting among those who share a language, a historic period, and/or a geographic ___location.” (Eisler et al. 2003). By Person we refer to the declarative parts of a person’s long-term memory where all kinds of knowledge, experiences and skills are stored and organized, and that person’s demographic variables, such as age, gender, education, etc. We also label the length of place-related stimulation and person’s momentary clothing as Situation.

Human response

Human response entails: Sensation (sensory unconscious detection of environmental stimulation/information); Perception (conscious interpretation and elaboration of sensory data), Cognition (how we learn, remember, and think about information); Emotion (states, processes, and expressions that convey qualities of affect, feeling, and mood); Adaptation (physiological regulation: a reduction in the responsiveness of a receptor as a result of repeated stimulation); Coping (psychological regulation: strategic ways for dealing with, for example, place-related stimulation); and Behaviour (activities, doings, reactions, movements).

An example of the place–moderator/mediator–human response link

In order to pedagogically exemplify the proposed place–moderator/mediator–human response link outlined in Fig. 1, imagine a summer day. The sky is clear (cloudlessness) and it is 22°C (air temperature). Some people are jogging (physical activity) and some are resting (physical activity) and chatting in groups (social activity) at a lawn in the central park (___location). This place is wide (openness) with a lot of vegetation (naturalness). Svea Svensson (“the subject”), a middle-aged (age) woman (gender) living in Sweden (Nordic culture) walks (behavior/physical activity) through the park (place), something she usually does on Saturdays (preference). She likes to walk in the park (belief/preference) because she finds pleasure in the nature and naturalness (attitude/expectation/emotion). Her sensory system is detecting the physical energy and the thermal conditions (thermal sensation). It converts this, for Svea Svensson unaware information (sensation), to the chemical energy which ultimately reaches her brain; that is, the short-term and long-term memory (cognition) where information is transformed into, for Svea Svensson, meaningful information (perception). She suddenly feels uncomfortable (thermal perception/comfort). It is too hot (thermal perception), she says to herself (cognition). Can I do something about it (perceived control), I wonder (cognition)? She then takes off (coping) the jacket (clothing) and by so adjusts (adaptation: physiological and psychological) to the momentary weather, warmness (thermal sensation/perception). I have been too long in the sun (length of exposure), she says to herself (cognition). Finally, she rests (behavior) in the shade under a tree because it is too sunny (cloudlessness/radiation) and warm (air temperature) in the open space. It is really beautiful here (perception) and I feel so calm and glad (emotion) being in this park under a sheltering tree, she ends.

Empirical study

Following Fig. 1 we investigated the impact of weather (place) and personal factors (moderator) on participants’ perceptual and emotional estimations (human response) of outdoor urban places in a Nordic city. This was done across four outdoor urban places and seasons in the city of Gothenburg (Göteborg), Sweden. Data from micrometeorological and psychological measurements were combined in a quasi-experimental design and analyzed by univariate (ANOVA) and multivariate (MANOVA) analyses of variance (see "Experimental Design" and "Results" below). In line with some earlier results (Nikolopoulou and Lykoudis 2006; Eliasson et al. 2007; Thorsson et al. 2007), air temperature, cloudlessness, and wind speed were included as independent variables (weather). Because of other results (Lawton et al. 1982; Franck 2002; Knez 2005), gender, age, and environmental attitude (open-air vs urban person) were also included as independent variables (personal factors). Given this, we predicted an influence of place (weather parameters) and moderator (personal factors) on participants’ perceptual and emotional evaluations (human response) of outdoor urban places, thus accounting for some of the unexplained variance between the objective and subjective estimations of outdoor urban places, as suggested by Nikolopoulou et al. (2001), Thorsson et al. (2004), and Nikolopoulou and Lykoudis (2006).

It must be noted that we have included the weather (defined as a part of a place in Fig. 1) both as a dependent (perceived/subjective) variable (How do you perceive the weather today?) and as an independent (objective) variable (levels of air temperature, cloudlessness and wind speed).

Study area and climate

The Gothenburg area (Göteborg, Sweden, 57°42’N, 11°58’E; population ca. 500 000) has a maritime climate with average air temperature of 6.8°C. The area has for the latitude relatively mild winters and cool summers, with average temperatures of −0.5°C in February and 16.3°C in July.

Participants

A total of 2,375 persons (47% men, 53% women) visiting one of the four outdoor urban places (a park, a square, a courtyard, or a waterfront plaza) located within the city of Gothenburg participated in this study. Their age (in years) fell in the ranges of: <36 (44%); 36–50 (23%); 51–70 (33%). In addition, 53% of the participants estimated themselves as urban and 47% as open-air persons (see below).

Micrometeorological measurements

A mobile meteorological station was used to measure the air temperature (Rotronic YA-100), wind speed (Gill Institute Ultrasonic) and short-wave radiation (Kipp and Zonen CM3). Air temperature and short-wave radiation were measured at 1.1 m, corresponding to the average height of the center of gravity for adults (Mayer and Höppe 1987). Wind speed was measured at 2 m height, and later recalculated down to 1.1 m using Sverdrup’s power law. Data were sampled every minute and stored in data loggers (Campbell CR 10). The meteorological station was placed at the most open point in each place so that the stations would be in the sun during the whole field survey.

Psychological measures

Structured interviews were carried out in each place. This instrument comprised questions about demographic variables, general and specific questions about current weather and place, and behaviors and attitudes related to place and person (Thorsson et al. 2007). Four questions from the questionnaire will be analyzed in the present study (see below). They are estimations of perceptual and emotional dimensions of weather and place, and of the urban versus open-air attitude in participants:

  1. 1.

    How do you perceive the weather today? (perception: perceived weather). Participants were asked to answer this question by responding to three 5-point scales: (1) calm–windy; (2) cold–warm; and (3) good–bad for outdoor activity (Thorsson et al. 2007).

  2. 2.

    How do you perceive the place right now? (perception: perceived place). Participants were asked to answer this question by responding to four 5-point scales: (1) ugly–beautiful; (2) unpleasant–pleasant; (3) windy–calm; and (4) cold–warm (Thorsson et al. 2007).

  3. 3.

    How do you feel right now in this place? (emotional and thermal perception). Participants were asked to answer this question by responding to four 5-point scales: (1) elated–bored; (2) glad–gloomy; (3) calm–nervous; (4) active–passive. These scales were derived from Knez and Hygge (2001). Participants were also asked to estimate their thermal comfort by responding to a 9-point scale; from very cold to very hot with the score 5 as comfortable (Matzarakis and Mayer 1996)

  4. 4.

    Urban versus open-air person attitude. (moderator: attitude/expectation). Participants’ urban versus open-air person attitude was measured on a 5-point scale: 1 (mostly urban person) to 5 (mostly open-air person) related to the question: “How much of an urban person (find pleasure in the street-life, the shops, the amusements of the city) and open-air person (find pleasure in the sea, the woods, the nature) are you?” (Knez 2005).

Experimental design

A non-equivalent comparison-group quasi-experimental design McGuigan (1983) was used, meaning that participants were not randomly assigned to different groups and that independent variables might have been confounded with some uncontrolled extraneous variables. In consequence and compared with a “true experiment” (Liebert and Liebert 1995), the inferences drawn about the causal relationships between independent and dependent variables might be considered to be weaker.

Independent variables

Three place variables (air temperature, cloudlessness, wind speed) and three moderators/personal factors (environmental attitude, gender, age) were used. Cloudlessness measured by the clearness index (CI) is defined as the ratio between a measured and a theoretical maximum incoming solar radiation for a specific time and ___location on the earth’s surface (Crawford and Duchon 1999). High values of CI (e.g., >0.75) represent clear sky conditions, while lower values represent more cloudy conditions. In order to get as many values as possible in the independent cells, categorizations of the independent variables were made pragmatically; that is, as few as possible independent variable categorizations with similar ranges were used:

  1. 1.

    Air temperature (T a ) was categorized into: (1) −8−0°C; (2) 0−8°C; (3) 8−16; °C; (4) 16−24°C.

  2. 2.

    Clearness index (CI) was categorized into: (1) 0–0.25; (2) 0.25–0.75; (3) 0.75–1.

  3. 3.

    Wind speed (W) was categorized into: (1) 0–1.5 m/s; (2) 1.5–3 m/s.

  4. 4.

    Environmental attitude was categorized into: (1) urban person; (2) open-air person. In line with Knez (2005) participants with scores lower than 3 (1, 2) were considered to be “urban persons” and those higher than 3 (4, 5) were considered to be “open-air persons”.

  5. 5.

    Gender: (1) men; (2) women.

  6. 6.

    Age was categorized into: (1) <36; (2) 36–50, (3) 51–70.

Dependent variables

Three questions comprising several scales (see "Psychological Measures", above) and measuring perceptual and emotional place-related assessment (human response): “How do you perceive the weather today?”; “How do you perceive the place right now?”; “How do you feel right now in this place?”.

Results

All data were subjected to univariate, ANOVA (for “thermal comfort” scale), or multivariate analyses of variance, MANOVAs (for dependent variables/questions: “How do you perceive the weather today?”; “How do you perceive the place right now?” and “How do you feel right now in this place?” comprising several scales) with six between-subject independent variables: 4 Ta × 3 CI × 2 W × 2 Environmental Attitude × 2 Gender × 3 Age. Replicating Knez and Thorsson (2006, 2008) results showed no significant influence of Gender. Thus, only significant results related to the other five independent variables will be reported. The analyses of variance used the method of unweighted means, given the unbalanced cell frequencies. All the 144 cells (2 attitudes × 3 age groups × 4 air temperatures × 3 CIs × 2 wind conditions) comprised a number of observations, and the number of observations in each cell across the three weather variables varied between 60 and 222 for environmental attitudes and between 52 and 274 for age groups.

Perceived weather

A multivariate significant main effect of Environmental Attitude on participants’ estimation of current weather was shown (Wilks’ Lambda = .99, F(3, 674) = 2.74, p  = ..042), associated with the cold-warm scale, F(1, 807) = 7.23, p = .007, showing that open-air persons estimated the current weather, independently of the combinations of the three weather parameters, as warmer than did urban persons (see Fig. 2).

Fig. 2
figure 2

Mean temperature (cold–warm scale) estimation of current weather as a function of environmental attitude (urban vs open-air persons)

Full size image

In addition and as can be seen in Fig. 3, a multivariate significant main effect of Age (Wilks’ Lambda = .98, F(6, 1,348) = 2.39, p = .027), associated with the calm–windy scale, F(2, 807) = 3.20, p = .041, showed that the youngest participants estimated the current weather as calmest, independently of the weather.

Fig. 3
figure 3

Mean wind (calm/windy scale) estimation of current weather as a function of age

Full size image

Multivariate significant main effects of CI (Wilks’ Lambda = .95, F(6, 1348) = 5.41, p = .000) and Ta (Wilks’ Lambda = .97, F(9, 1,640) = 28.04, p = .000), associated with the good–bad for outdoor activity scale (CI, F(1, 807) = 7.40, p = .001 and Ta, F(3, 807) = 6.76, p = .000) showed that the participants estimated the current weather as better for outdoor activity as a function of higher CI and Ta respectively (see Figs. 4 and 5).

Fig. 4
figure 4

Mean good–bad for outdoor activity estimation of current weather as a function of clearness index (CI)

Full size image
Fig. 5
figure 5

Mean good–bad for outdoor activity estimation of current weather as a function of air temperature (Ta)

Full size image

Furthermore, a tendency to a significant multivariate interaction between W and Environmental attitude on participants’ estimations of current weather (Wilks’ Lambda = .99, F(3, 674) = 2.26, p = .080), associated with the cold-warm scale, F(1, 807) = 6.05, p = .014, showed that only the open-air persons were sensitive enough to estimate a difference in current weather as related to the two W categories of “0–1.5 m/s” and “1.5–3 m/s”; meaning that the current weather was estimated as colder only by the open-air persons, as a result of stronger W (see Fig. 6).

Fig. 6
figure 6

Mean temperature (cold–warm scale) estimation of outdoor urban places as a function of wind speed (m/s) and environmental attitude (urban vs open-air persons)

Full size image

In sum, the results have shown that the current weather was estimated as most good for outdoor activity as a function of high air temperature and clear sky. The age of and the environmental attitude in participants also influenced the current weather assessments. The youngest participants were shown to estimate the current weather as calmest and the open-air persons compared to urban persons were shown to estimate the current weather as warmer, independently of the weather parameter combinations. Finally, and compared to the urban persons, the open-air persons were shown to be more sensitive to wind speed variations; meaning that only they perceived the wind speed differences between the wind speed categories of “0–1.5 m/s” and “1.5–3 m/s”.

Perceived place

Two significant multivariate main effects of CI (Wilks’ Lambda = .97, F(8, 13,40) = 2.83, p = .004) and W (Wilks’ Lambda = .94, F(4, 670) = 10.35, p = .000) on participants’ estimations of the outdoor urban places, associated with cold–warm, F(2, 804) = 5.96, p = .003, and windy-calm scales, F(1, 804) = 35.25, p = .000, showed that the outdoor urban places were estimated as warmer when CI rose and windier when W increased (see Figs. 7 and 8).

Fig. 7
figure 7

Mean temperature (cold–warm scale) estimation of outdoor urban places as a function of clearness index (CI)

Full size image
Fig. 8
figure 8

Mean wind (windy–calm scale) estimation of outdoor urban places as a function of wind speed (m/s)

Full size image

Thus, according to the results obtained, the participants perceived the outdoor urban places as warmer when the sky was clear and windier with a stronger wind.

Place-related emotions and thermal comfort

Due to some practical flows in measuring thermal comfort, it must be noted that data regarding this measure were obtained only for two Ta conditions: “8–16°C” and “16–24°C”. A tendency to a significant multivariate interaction between CI and Environmental Attitude on participants’ place-related emotions (Wilks’ Lambda = .98, F(8, 1,342) = 1.70, p = .094) associated with calm-nervous scale, F(2, 805) = 3.11, p = .045, showed that urban persons felt most calm and open-air persons least calm in the CI “0.75=1” condition (see Fig. 9). A significant multivariate interaction between CI and Ta on participants’ place-related emotions (Wilks’ Lambda = .95, F(24, 2342) = 1.57, p = .039) associated with glad-gloomy scale, F(6, 805) = 4.28, p = .000, showed that participants felt most glad when the weather conditions of CI “0.75–1” and Ta “16–24°C” were combined and least glad when the weather conditions of CI “0–0.25” and Ta “0–8°C” were present (see Fig. 10).

Fig. 9
figure 9

Mean calm–nervous feelings in participants as a function of clearness index (CI) and environmental attitude (urban vs open-air person)

Full size image
Fig. 10
figure 10

Mean glad–gloomy feelings in participants as a function of air temperature (Ta) and clearness index (CI)

Full size image

Two main effects of Environmental Attitude, F(1, 403) = 6.44, p = .012, and Ta, F(1, 403) = 5.20, p = .023, on participants’ thermal comfort in outdoor urban places showed that urban persons felt thermally more comfortable than did open-air persons (urban persons M = 5.0, SE = .10; open-air persons M = 5.4, SE = .10) and that all participants felt thermally more comfortable in Ta “8–16°C” than in Ta “16–24°C” condition (Ta “8–16°C” M = 4.98, SE = .10; Ta “16–24°C” M = 5.4, SE = .11). Finally, a significant interaction between Environmental Attitude and Age, F(2, 403) = 2.96, p = .053, showed that urban persons’ thermal comfort decreased (from 5.20 to 4.82 on a 9-point scale) with age and that open-air persons’ thermal comfort increased (from 5.16 to 5.47 on a 9-point scale) with age (see Fig. 11).

Fig. 11
figure 11

Mean thermal comfort in participants as a function of age and environmental attitude (urban vs open-air person)

Full size image

In sum, and concerning the place-related emotions, the results obtained showed that the urban persons felt calmer when the sky was clear and that the open-air persons, on the other hand, felt most calm when the sky was cloudier. However, all participants felt most glad when the air temperature was high and the sky was clear, and least glad when the weather conditions comprised low air temperature and cloudy sky. As regards the thermal comfort, the participants felt most comfortable in the Ta “8–16°C” condition. Interestingly and independently of the weather conditions, the urban compared to the open-air persons were shown to feel thermally more comfortable in the outdoor urban places. Finally it was indicated that the urban and the open-air persons’ thermal comfort might vary, in an opposite direction, as a function of age.

Discussion

The aim of this article was to illuminate the psychological (individual) mechanisms involved in outdoor place and weather assessment in order to broaden previous work on the thermal comfort indices as well as on research addressing the significance of weather for the use and design of urban places. This was done by clarifying the human information processing and how mental representations, such as schemata, stored in long-term memory might influence place-related responses. This reasoning was also conceptualized in a model, tentatively proposing “direct” and “indirect” links of influence in a place–human relationship, which was subsequently tested by an empirical study on the impact of weather (place) and personal factors (mediator) on perceptual and emotional estimations (human response) of outdoor urban places.

Concerning “direct influence” (see Fig. 1), the participants were shown to perceive the weather as best for outdoor activity when the air temperature was high and the sky was clear. The outdoor urban places were also perceived as warmer when the sky was clear. The influence of air temperature and cloudlessness on environmental assessment and activity has previously been demonstrated by Nikolopoulou and Lykoudis (2006) and Eliasson et al. (2007). Furthermore, the participants perceived the outdoor urban places as windier when the wind speed was stronger, which reveals human ability to discriminate wind speed variations (Cohen et al. 1979; Bell et al. 2001). Concerning the place-related emotions, participants felt most glad when the air temperature was high and the sky was clear; and least glad when the air temperature was lower and the sky was cloudier. This indicates a relation between emotions and air temperature and cloudlessness which is, generally, in line with Cunningham (1979) showing that sunlight might lead to a positive mood.

Concerning “indirect influence” (see Fig. 1), and in line with Knez and Thorsson (2006, 2008), no effects of gender were shown. However, regarding the participants’ attitude/expectation (open-air vs urban person) and age, several significant results were reported. Independently of the weather conditions, the open-air persons compared to the urban persons were shown to estimate the current weather as warmer and were shown to be more sensitive to the wind speed variations. However, independently of the weather conditions, the urban persons compared to the open-air persons were shown to feel thermally more comfortable in the outdoor urban places. They also felt most calm when the sky was clear, compared to open-air persons who felt alike when the sky was cloudier. In line with some previous findings (Brewer and Treyens 1981; Knez 2005; Knez and Thorsson 2006, 2008), all this indicates an influence of the moderator/personal factor, environmental attitude: a place-related schema stored in the long-term memory that significantly guided and interpreted the open-air and the urban persons’ different experiences of and expectations towards the weather and the outdoor urban places. Furthermore, regarding the participants’ age, it was shown that youngest compared to middle-aged and older participants perceived the current weather as calmer. In line with Lawton et al. (1982), this suggests that age might also be an important moderator/personal factor in environmental assessment. What accounts for this result? Age as a demographic variable, on an individual level of explanation, can also be treated as a type of an attitude/schema comprising varying age-related experiences of and expectations towards the weather and the outdoor urban places.

Conclusions

Taken together and in line with the predictions, we have found significant influences of especially air temperature, cloudlessness, environmental attitude, and age on participants’ perceptual and emotional estimations of outdoor urban places. All this is a modest, yet significant, step towards an understanding of the psychology of outdoor place and weather assessment that future research on the thermal climate indices as well as on the significance of climate for the use and design of urban places ought to consider. Finally, we have to underline that the suggested model is a conceptual one, meaning that the model, in its present form, is vague. This means that future research must clarify in some more detail relationships among model components and formulate specific hypotheses regarding the processes of moderation and/or mediation.