Comment

Comment by Margaret Hsing on February 17, 2024 at 4:49pm

These researchers examined the effects of an office make-over when a company moved to a new office building. The employees in the new office were given individual control of the temperature, lighting, air quality, and acoustic conditions where they were work ing.

Productivity increased by approximately 15% in the new building. When the individual control of the ambi ent multisensory environment was disabled in the new building, performance fell by around 2% instead. Trying to balance the influence of each of the senses is one of the aims of Finnish architect Juhani Pallasmaa, whose name we have come across at several points already in this text.

As Steven Holl notes in the preface to Pallas maa’s The eyes of the skin: “I have experienced the archi tecture of Juhani Pallasmaa, … The way spaces feel, the sound and smell of these places, has equal weight to the Fig. 8 The Ira Keller Fountain, Portland Oregon. According to Pallasmaa (2011), p. 596) this is “An architecture for all the senses including the kinaesthetic and olfactory senses.”

Once again, the auditory element is provided by the sound of falling water way things look.” (Pallasmaa, 1996, p. 7). One example of multisensory architectural design to which Juhani Pal lasmaa draws attention in several of his writings is the Ira Keller Fountain, Portland Oregon (see Fig. 8). On the multisensory integration of atmospheric/ environmental cues To date, only a relatively small number of studies have directly studied the influence of combined ambient/at mospheric cues on people’s perception, feelings, and/or behaviour. Mattila and Wirtz (2001) conducted one of the first sensory marketing studies to be published in this area.

These researchers manipulated the olfactory environment (no scent, a low-arousal scent (lavender), or a high-arousal scent (grapefruit)) while simultan eously manipulating the presence of music (no music, low-arousal music, or high-arousal music). When the scent and music were congruent in terms of their arousal potential, the customers rated the store envir onment more positively, exhibited higher levels of ap proach and impulse-buying behaviour, and expressed more satisfaction.

There is, though, always a very real danger of sensory overload if the combined multisen sory input becomes too stimulating (see Malhotra, 1984; Simmel, 1995). Meanwhile, in another representative field study, Sayin et al. (2015) investigated the impact of presenting ambi ent soundscapes in an underground car park in Paris. In particular, they assessed the effects of introducing west ern European birdsong or classical instrumental music by Albinoni to the three normally silent stairwells used by members of the general public when exiting the car park. A total of 77 drivers were asked about their feel ings on their way out.

Birdsong was found to work best in terms of enhancing the perceived safety of the situation- in this case by around 6%. This despite the fact that all of those who were quizzed realized that the sounds that they had heard were coming from loud speakers.12 In an accompanying series of laboratory studies, Sayin et al.’s participants were shown a 60-s first-person perspective video that had been taken in the same Paris car park, or else a short video of someone walking through a metro station in Istanbul.

Comment by Margaret Hsing on February 16, 2024 at 7:10am

Once again, participants were asked about how safe it felt, about perceived social presence, and about their willingness to purchase a monthly metro pass. Even under these some what contrived experimental conditions, the presence of an ambient soundscape once again increased perceived safety as well as the participants’ self-reported intention to purchase a season ticket.

It was, though, the sound of people singing Alleluia that proved most effective in terms of enhancing perceived safety amongst those watching the videos.13 It is, however, worth bearing in mind here that many of the key results reported in this study were only borderline significant.

As such, adequately-powered repli cation would be a good idea before too much weight is given to these intriguing findings. Recently, Ba and Kang (2019) documented crossmodal interactions between ambient sound and smell in a laboratory study that was designed to capture the sensory cues that might be encountered in a typical urban environment.

These researchers decided to pair the sounds of birds, conversation, and traffic, with the smells of flowers (lilac, osmanthus), coffee, or bread, at one of three levels (low, medium, or high) in each modality. A complex array of in teractions was observed, with increasing stimulus intensity sometimes enhancing the participants’ comfort ratings, while sometimes leading to a negative response instead. While Ba and Kang’s results defy any simple synopsis, given the complex pattern of results reported, their find ings nevertheless clearly suggest that sound and scent interact in terms of influencing people’s evaluation of urban design.

The colour of the ambient lighting in an indoor envir onment has also been shown to influence the perceived ambient temperature and thermal comfort of an envir onment (e.g., Candas & Dufour, 2005; Tsushima, et al., 2020; Winzen, Albers, & Marggraf-Micheel, 2014). For instance, in one representative study, Winzen and col leagues reported that illuminating a simulated aircraft cabin in warm yellow vs. cool blue-coloured lighting 12This response is very different from the aesthetic disappointment, or even disgust, felt by the man once hypothetically described by the philosopher Immanuel Kant who was very much enjoying listening to a nightingale’s song until realizing that he was listening to a mechanical imitation instead (Kant, 2000). 13

The owner of the car park did not like the sound of this particular sonic intervention, meaning that the researchers were unable to try it out in the field. exerted a significant influence over people’s self-reported thermal comfort. The participants rated the environment as feeling significantly warmer under the warm (as com pared to the cool) lighting colour. One can only really make sense of such findings from a multisensory per spective (see Spence, 2020a, for a review). Taken together, then, the results of the representative selection of studies reported in this section demonstrate that our perception of, and/or response to, multisensory environments are undoubtedly influenced by the com bined influence of environmental/atmospheric cues in different sensory modalities.

So, in contrast to the quote from Mattila and Wirtz (2001) that we came across a few pages ago, there is now a growing body of empirical research out there demonstrating that atmospheric cues presented in different sensory modalities, such as music, scents, and visual stimuli combine to influence how alerting, or pleasant, a particular environment, or stimulus (such as, for example, a work of art), is rated as being (e.g., Banks, Ng, & Jones-Gotman, 2012; Battacharya & Lindsen, 2016).

Comment by Margaret Hsing on February 15, 2024 at 10:01am

Sensory congruency In their book, Spaces speak, are you listening?, Blesser and Salter draw the reader’s attention to the importance of audiovisual congruency in architectural design. They write that: “Aural architecture, with its own beauty, aes thetics, and symbolism, parallels visual architecture. Vis ual and aural meanings often align and reinforce each other. For example, the visual vastness of a cathedral communicates through the eyes, while its enveloping re verberation communicates through the ears.” (Blesser & Salter, 2007, p. 3). However, they also draw attention to the incongruency that one experiences sometimes: “Al though we expect the visual and aural experience of a space to be mutually supportive, this is not always the case. Consider dining at an expensive restaurant whose decorations evoke a sense of relaxed and pampered ele gance, but whose reverberating clatter produces stress, anxiety, isolation, and psychological tension, undermin ing the possibility of easy social exchange.

The visual and aural attributes produce a conflicting response.” (Blesser & Salter, 2007, p. 3). Regardless of whether atmospheric/environmental sen sory cues are integrated or not, one general principle underpinning our response to multisensory combina tions of environmental cues is that those combinations of stimuli that are “congruent” (whatever that term means in this context) will tend to be processed more fluently, and hence be liked more, than those combina tions that are deemed incongruent, and hence will often prove more difficult, and effortful, to process (Reber, 2012; Reber, Schwarz, & Winkielman, 2004; Reber,
Winkielman, & Schwartz, 1998; Winkielman, Schwarz, Fazendeiro, & Reber, 2003; Winkielman, Ziembowicz, & Nowak, 2015).14 Indeed, it was the putative sensory incongruency between a relaxing slow-tempo music and arousing citrus scent that was put forward as a possible explanation for why Morrin and Chebat (2005) found that adding scent and sound in the setting of the shop ping mall reduced unplanned purchases as compared to either of the unisensory interventions amongst almost 800 shoppers in one North American Mall (see Fig. 9). Congruency can, of course, be defined at multiple levels. For instance, as we have seen already in this sec tion, sensory cues may be more or less congruent in terms of their arousal/relaxation potential (e.g., Hom burg, Imschloss, & Kühnl, 2012; Mattila & Wirtz, 2001). Mahvash (2007, pp. 56–57) talks about the use of con gruent cues to convey the notion of coolness: “… the Persian garden with its patterns of light and shadow, reflecting pools, gurgling fountains, scents of flowers and fruits, and gentle cool breezes 'offers an amazing rich ness of variety of sensory experiences which all serve to reinforce the pervasive sense of coolness'.” However, dif ferent sensory inputs may also be deemed congruent or not in terms of their artistic style (see Hasenfus, Martin dale, & Birnbaum, 1983; Muecke & Zach, 2007; cf. Her sey, 2000, pp. 37–41).

Comment by Margaret Hsing on February 14, 2024 at 4:10pm

It was stylistic congruency that was manipulated in a couple of experiments, conducted 14At the same time, however, one might consider how marble, one of the most highly prized building materials is in some sense incongruent, given the rich textured patterning of the veined appearance of the surface is typically perfectly smooth to the touch. both online and in the laboratory by Siefkes and Arielli (2015).

These researchers had their participants expli citly concentrate on and evaluate the style of the buildings shown in one of two architectural styles (baroque or modern- a short video showing five baroque build ings; there were also a short video, focusing on five mod ern buildings instead). Their results revealed that the buildings were rated as looking more balanced, more co herent, and to a certain degree, more complete,15 when viewed while listening to music that was congruent (e.g., baroque architecture with baroque music- specifically Georg Philipp Telemann’s, Concerto Grosso in D major, TWV 54:D3 (1716)) rather than incongruent (e.g., bar oque architecture with Philip Glass track from the soundtrack to the movie Koyaanisqatsi).

Before moving on, though, it is worth noting that in this study, as in many of the other studies reported in this section, there is a possibility that the design of the experiments themselves may have resulted in the partici pants concerned paying rather more attention to the at mospheric/environmental cues (and possibly also their congruency) than is normally likely to be the case when, as was mentioned earlier, the architecture itself fades into the background.

Ecological validity may, in other words, have been compromised to a certain degree. One of the other examples of incongruency that one often comes across is linked to the growing interest in biophilic design. As Pallasmaa (1996, p. 41) notes: “A walk through a forest is invigorating and healing due to 15These were the anchors on three of the bipolar semantic differential scales used in this study.

the constant interaction of all sense modalities; Bachelard speaks of ‘the polyphony of the senses’. The eye collaborates with the body and the other senses. One’s sense of reality is strengthened and articulated by this constant interaction. Architecture is essentially an extension of nature into the man-made realm …”16 No wonder, then, that many designers have been exploring the benefits of bringing elements of nature into interior spaces in order to boost the occupants’ mood and aid relaxation (Spence, 2021).

However, one has to ask whether the benefits of adding the sounds of a tropical rainforest to a space such as the shopping area of Glasgow airport, say (Treasure, 2007), really outweigh the cognitive dissonance likely elicited by hearing such sounds in such an incongruous setting? Similarly, a jungle soundscape was incorporated into the children’s section of Harrods London Department store a few years ago (Harrods’ Toy Kingdom- The Sound Agency | Sound Branding” https:// www.youtube.com/watch?v=EVUUG6VvFKQ).

Nature soundscapes have also been introduced into Audi car salesrooms, not to mention BP petrol station toilet facilities (Bashford, 2010;Treasure, 2007). It is worth noting here that given the important role that congruency has been shown to play at the level of multisensory object/ event perception, there is currently a stark paucity of research that has systematically investigated the relevance/ importance of congruency at the level of multisensory ambient, or environmental, cues. As the quotes earlier in this section make clear, it is something to which some architects are undoubtedly sensitive, and on which they already have an opinion. Yet the relevant underpinning research still needs to be conducted.

Comment by Margaret Hsing on February 13, 2024 at 8:02pm

Ultimately, therefore, while the congruency of atmos pheric/environmental cues can be defined in various ways, and while incongruency is normally negatively valenced (because it is hard to process),17 issues of (in)congruency may often simply not be an issue for the occupants of specific environments. This may either be because the latter simply do not pay attention to the at
mospheric/environmental cues (and hence do not register their incongruency) and/or because they have no reason to believe that the stimuli should be combined in the first place.16

The value of connecting with nature in architectural design practice was stressed by an advertorial for an arctic hideaway that suggests that: “True luxury today is connecting with nature and feeling that your senses work again” as appeared in an article in Blue Wings magazine (December 2019, p. 38). 17

It should, though, be remembered, that sometimes incongruency may be precisely what is wanted. Just take the following quote regarding the crossmodal contrast of thermal heat combined with
visual coolness from Japan as but one example: “In the summer the householder likes to hang a picture of a waterfall, a mountain stream, or similar view in the Tokonama and enjoy in its contemplation a feeling of coolness.” (Tetsuro, 1955, p. 16).

Sensory dominance

One common feature of configurations of multisensory stimuli that are in some sense incongruent is sensory dominance. And very often, under laboratory conditions, this tends to be vision that dominates (e.g., Hutmacher, 2019; Meijer et al., 2019; Posner et al., 1976). Under conditions of multisensory conflict, the normally more reliable sense sometimes completely dominates the
experience of the other senses, as when wine experts can be tricked into thinking that they are drinking red or rosé wine simply by adding some red food dye to white wine (Wang & Spence, 2019). Similarly, people’s assess ment of building materials has also been shown to be dominated by the visual rather than by the feel (Wastiels, Schifferstein, Wouters, & Heylighen, 2013; see also Karana, 2010).

At the same time, however, while we are largely visually dominant, the other senses can also sometimes drive our behaviour. For instance, according to an article that appeared in the Wall Street Journal, many people will apparently refuse to check in to a hotel if there is funny smell in the lobby (Pacelle, 1992). Such admittedly anecdotal observations, were they to be backed up by robust empirical data, would then support the notion that olfactory atmospheric cues can, at least under
certain conditions, also dominate in terms of determining our approach-avoidance behaviour. Mean
while, a growing number of diners have also reported how they will sometimes leave a restaurant if the noise is too loud (see Spence, 2014, for a review; Wagner, 2018), resonating with the quote from Blesser and Salter (2007) that we came across a little earlier.

One other potentially important issue to bear in mind here concerns the “assumption of unity”, or
coupling/binding priors that constitute an important factor modulating the extent of crossmodal binding in the case of multisensory object/event perception, according to the literature on the currently popular Bayesian causal inference (see Chen & Spence, 2017; Rohe, Ehlis,&Noppeney, 2019, for reviews). Coupling priors can be thought of as the internalized long-term statistics of the environment (e.g., Girshick, Landy, & Simoncelli, 2011).

Comment by Margaret Hsing on February 12, 2024 at 5:12pm

Does it, I wonder, make sense to suggest that we have such priors concerning the unification of environmental/atmospheric cues? Or might it be, perhaps, that in a context in which we are regularly exposed to incongruent environmental/atmospheric multisensory cues- just think of how music is played from loudspeakers without any associated visual referent- that out priors concerning whether to integrate what we see, hear, smell, and feel will necessarily be related, in any meaningful sense, may well be reduced substantially.

See Badde Navarro, and Landy (2020) and Gau and Noppeney  (2016) on the role of context in the strength of the  common-source priors multisensory binding.

Hence, no matter whether one wants to create a tranquil space (Pheasant, Horoshenkov, Watts, & Barret, 2008)or one that arouses (Mattila & Wirtz, 2001), the senses interact as they do in various other configurations and situations (e.g., Jahncke, Eriksson, & Naula, 2015; Jiang,  Masullo, & Maffei, 2016). There are, in fact, numerous examples where the senses have been shown to interact in  the experience and rating of urban environments (e.g., Ba &Kang,2019; Van Renterghem & Botteldooren, 2016).

Crossmodal correspondences in architectural design practice The field of synaesthetic design has grown rapidly in  recent years (e.g., Haverkamp, 2014; Merter, 2017;  Spence, 2012b). According to architectural historian, Alberto Pérez-Gómez, mentioned earlier, the Philips Pavilion designed by Le Corbusier for the 1958 Brussels world’s fair (Fig. 10) attempted to deliver a multisensory experience, or atmosphere by means of “forced” synaesthesia (Pérez-Gómez, 2016,p.19).18

The interior audiovisual environment was mostly designed by Le Corbusier and Iannis Xenakis (see Sterken, 2007). From those descriptions that have survived there were many  coloured lights and projections and a looping soundscape that was responsive to people’s  ovement through the  space (Lootsma, 1998; Muecke & Zach, 2007). 

18 Though Pérez-Gómez (2016, p. 65) seems to be using a rather unconventional definition of synaesthesia, as a little later in his otherwise excellent work, he defines perceptual synaesthesia as “the integrated sensory modalities”, Pérez-Gómez (2016, p. 65). The  majority of cognitive neuroscientists would, I presume, take this as a  definition of multisensory perception, rather than synaesthesia. Synaesthesia, note, is typically defined as the automatic elicitation  of an idiosyncratic concurrent, not normally experienced, in response  to the presence of an inducing stimulus (Grossenbacher & Lovelace,  2001).

True to his oculocentric approach, mentioned at the start of this piece, Le Corbusier apparently concentrated  on the visual aspects of the “Poème Electronique”, the multimedia show that was projected inside the pavilion.

Meanwhile, his site manager, Iannis Xenakis created “Concret PH”- the soundscape, broadcast over 300 loudspeakers, that accompanied it. It is, though, unclear how much connection there actually was between the auditory and visual components of this multimedia presentation. The notion of parallel, but unconnected, stimulation to eye and ear comes through in Xenakis’ quote that: “we are capable of speaking two languages at the same time.

Comment by Margaret Hsing on February 11, 2024 at 3:31pm

One is addressed to the eyes, the other to the ears.” (Varga,  1996,p.114).Moreover, inhis laterwork(e.g.,Polytopes),  Xenakis pursued the idea of creating a total dissociation be tween visual and aural perception in large abstract sound and light installations (Sterken, 2007, p. 33).

 At several points throughout his book Pérez-Gómez (2016), stresses the importance of “synaesthesia” to architecture, without, unfortunately, ever really quite defining what he means by the term. All one finds are quotes such as the following: “primordial synesthetic perception”,  p. 11;  “perception is primordially  synesthetic”, p. 20; “synaesthesia as the primary modality  of human perception”, p. 71. Pérez-Gómez (2016, p.  149) draws heavily on Merleau-Ponty’s (1962, p. 235) Phenomenology of Perception, quoting lines such as:

 “The senses translate each other without any need of an interpreter, they are mutually comprehensible without the intervention of any idea.” A few pages later he cites Heidegger “truths as correspondence” (Pérez-Gómez,  2016, p. 162). This does, though, sound more like a de scription of the ubiquitous crossmodal correspondences  (Marks, 1978; Spence, 2011) than necessarily fitting with  contemporary definitions of synaesthesia, though the distinction between the two phenomena admittedly remains fiercely contested (e.g., Deroy & Spence, 2013; Sathian & Ramachandran, 2020). Abath (2017) has done a great job of highlighting the confusion linked to Merleau-Ponty’s incoherent use of the term synaesthesia, that has, in turn, gone on to “infect” the writings of other architectural theorists, such as Pérez-Gómez (2016).

Talking of synaesthetic design may then be something  of a misnomer (Spence, 2015), the fundamental idea here is to base one’s design decisions on the sometimes surprising connections between the senses that we all share, such as, for example, between high-pitched sounds and small, light, fast-moving objects (e.g.,  Spence, 2011, 2012a). It is important to highlight the fact  that while these crossmodal correspondences are often confused with synaesthesia, they actually constitute a superficially similar, but fundamentally quite different empirical phenomenon (see Deroy & Spence, 2013).

We have already come across a number of examples of crossmodal correspondences being incorporated,  knowingly or otherwise, in design decisions. Just think about the use of temperature-hue correspondences  (Tsushima et al., 2020; see Spence, 2020a, for a review).

The lightness-elevation mapping (crossmodal correspondence) might also prove useful from a design perspective (Sunaga, Park, & Spence, 2016). And colour taste and sound-taste correspondences have already been incorporated into the design of multisensory experiential spaces (e.g., Spence et al., 2014; see also Adams &  Doucé, 2017; Adams & Vanrie, 2018). Once one accepts  the importance of crossmodal correspondences to environmental design, then this represents an additional level  at which sensory atmospheric cues may be judged as  congruent (e.g., see Spence et al., 2014). One of the important questions that remains for future research,  though, is to determine whether there may be a priority of one kind of cross modal congruency over others when they are manipulated simultaneously.

Comment by Margaret Hsing on February 8, 2024 at 5:39pm

Conclusions

While it would seem unrealistic that the dominance, or hegemony (Levin, 1993), of the visual will be overturned any time soon, that does not mean that we should not do our best to challenge it. As critic David Michael Levin puts it: “I think it is appropriate to challenge the hegemony of vision– the ocular-centrism of our culture.

And I think we need to examine very critically the character of vision that predominates today in our world. We urgently need a diagnosis of the psychosocial path ology of everyday seeing– and a critical understanding of ourselves as visionary beings.” (Levin, 1993, p. 205).

While not specifically talking about architecture, what we can all do is to adopt a more multisensory perspective and be more sensitive to the way in which the senses interact, be it in architecture or in any other as pect of our everyday experiences.

By designing experiences that congruently engage more of the senses we may be better able to enhance the quality of life while at the same time also creating more immersive, engaging, and memorable multisensory experiences (Bloomer & Moore, 1977; Gallace & Spence, 2014; Garg, 2019; Spence, 2021; Ward, 2014). Stein and Meredith (1993, p. xi), two of the foremost multisensory
neuroscientists of the last quarter century, summarized this idea when they suggesting in the preface to their in fluential volume The merging of the senses that: “The in tegration of inputs from different sensory modalities not only transforms some of their individual characteristics, but does so in ways that can enhance the quality of life.

Integrated sensory inputs produce far richer experiences than would be predicted from their simple coexistence or the linear sum of their individual products.” There is growing interest across many fields of endeavour in design that moves beyond this one dominant, or perhaps even overpowering, sense (Lupton & Lipps, 2018). The aim is increasingly to design for experience rather than merely for appearance. At the same time, however, it is also important to note that progress has been slow in translating the insights from the academic field of multisensory research to the world of architec
tural design practice, as noted by licensed architect Joy Monice Malnar when writing about her disappointment with the entries at the 2015 Chicago Architecture Biennial.

Comment by Margaret Hsing on February 8, 2024 at 5:39pm

There, she writes: “So, where are we? What is the current state of the art? Sadly, the current research on multisensory environments appearing in journals such as The Senses & Society does not appear to be impacting artists and architects participating in the Chicago Biennial. Nor are the discoveries in neuroscience offering new information about how the brain relates to the physical environment.” (Malnar, 2017, p. 153).19 At the same time, however, the adverts for at least one new residential development in Barcelona promising residents the benefits of “Sensory living” (The New York Times International Edition in 2019, August 31–Septem ber 1, p. 13), suggests that at least some architects/de signers are starting to realize the benefits of engaging their clients’/customers’ senses. The advert promised that the newly purchased apartment would “provoke their senses”.

Ultimately, it is to be hoped that as the growing awareness of the multisensory nature of human perception continues to spread beyond the academic community, those working in the field of architectural design practice will increasingly start to incorporate the multisensory perspective into their work; and, by so doing, promote the development of buildings and urban spaces that do a better job of promoting our social, cognitive, and emotional well-being.

(Source: Senses of place: architectural design for the multisensory mind by Charles Spence; in Cognitive Research: Principles and Implications (2020) 5:46 Keywords: Multisensory perception, Architecture, The senses, Crossmodal correspondences)

Related:

地方感性

愛懇雲端藝廊：設計故事館

Comment by Margaret Hsing on December 19, 2023 at 6:33pm

陈明发〈从“意识流”谈起〉

“意识流”理论易懂难精。

除了伍尔夫的著作，阅读弗拉基米尔·纳博科夫的《说吧，记忆：自传追述》也带给我很大的乐趣。此书部分地采纳了意识流。

意识流最有名的小说之一《尤利西斯》，买了许久一直没好好读，很对不起乔伊思。另一部意识流大书普鲁斯特的《追忆逝水年华》，我偷工减料地在一些选集中，当散文篇章读过一小部分，说起来很不成样子。

但在哲学方面。柏格森（1927年诺贝尔文学奖得主）对我影响很大，因为他，我上溯维柯（“诗性智慧”历史发展观），下访克里斯蒂瓦（“诗性语言”革命），思想起来心里比较踏实些。

“意识流”得益于19世纪心理学家威廉·詹姆斯的学说，但后来被归入“意识流”的文学家，明显也采纳了20世纪佛洛伊德、荣格、拉康诸子的精神分析理论。

对中国意识流文学作品我不熟悉，我近年相对喜欢的小说家如李佩甫（《无边无际的早晨》等）、冯苓植（《与死共舞—“鞭杆”的故事》等）的创作。二人把意识流与“微魔幻”做了令人期许的混融。

我个人的心智之旅，许多年来依靠着存在主义先驱海德格尔的哲学，近年有柏格森诸子加持，期待文创路上有新领悟。再读李佩甫和冯苓植等中文作家的努力结果，诗性实践在汉字文创中无疑是可能的。（19.12.2023）

• ‹ Previous
• 1
• …
• 3
• 4
• 5
• 6
• 7
• Next ›
• Page